Meng-Chieh Lee, Qi Zhu, Costas Mavromatis, Zhen Han, Soji Adeshina, Vassilis N. Ioannidis, Huzefa Rangwala, Christos Faloutsos. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

Tabular data within enterprises or open data repositories provide a huge opportunity for feature augmentation. Using these data sources to augment training data often boosts model performance, which is crucial in data-centric AutoML systems. Recent works on automatic feature augmentation have limited capabilities in utilizing useful features that cannot be joined with the base table without connecting through intermediate tables. We present Featpilot, a novel framework that explores and integrates high-quality features in tabular data for ML models. Featpilot evaluates a candidate feature from two aspects: (1) the efficacy of a join path connecting the feature to the base table and (2) the intrinsic value of a feature towards an ML task. Featpilot efficiently identifies high-quality features and their optimized join paths to augment the base table. Our experimental results show that Featpilot achieves up to a 10.27% improvement in ML model performance compared to state-of-the-art solutions across six public datasets.

Can integrating spectral and curvature signals unlock new potential in graph representation learning? Non-Euclidean geometries, particularly Riemannian manifolds such as hyperbolic (negative curvature) and spherical (positive curvature), offer powerful inductive biases for embedding complex graph structures like scale-free, hierarchical, and cyclic patterns. Meanwhile, spectral filtering excels at processing signal variations across graphs, making it effective in homophilic and heterophilic settings. Leveraging both can significantly enhance the learned representations. To this end, we propose Spectro-Riemannian Graph Neural Networks (CUSP) - the first graph representation learning paradigm that unifies both CUrvature (geometric) and SPectral insights. CUSP is a mixed-curvature spectral GNN that learns spectral filters to optimize node embeddings in products of constant curvature manifolds (hyperbolic, spherical, and Euclidean). Specifically, CUSP introduces three novel components: (a) Cusp Laplacian, an extension of the traditional graph Laplacian based on Ollivier-Ricci curvature, designed to capture the curvature signals better; (b) Cusp Filtering, which employs multiple Riemannian graph filters to obtain cues from various bands in the eigenspectrum; and (c) Cusp Pooling, a hierarchical attention mechanism combined with a curvature-based positional encoding to assess the relative importance of differently curved substructures in our graph. Empirical evaluation across eight homophilic and heterophilic datasets demonstrates the superiority of CUSP in node classification and link prediction tasks, with a gain of up to 5.3\% over state-of-the-art models.

Does the intrinsic curvature of complex networks hold the key to unveiling graph anomalies that conventional approaches overlook? Reconstruction-based graph anomaly detection (GAD) methods overlook such geometric outliers, focusing only on structural and attribute-level anomalies. To this end, we propose CurvGAD - a mixed-curvature graph autoencoder that introduces the notion of curvature-based geometric anomalies. CurvGAD introduces two parallel pipelines for enhanced anomaly interpretability: (1) Curvature-equivariant geometry reconstruction, which focuses exclusively on reconstructing the edge curvatures using a mixed-curvature, Riemannian encoder and Gaussian kernel-based decoder; and (2) Curvature-invariant structure and attribute reconstruction, which decouples structural and attribute anomalies from geometric irregularities by regularizing graph curvature under discrete Ollivier-Ricci flow, thereby isolating the non-geometric anomalies. By leveraging curvature, CurvGAD refines the existing anomaly classifications and identifies new curvature-driven anomalies. Extensive experimentation over 10 real-world datasets (both homophilic and heterophilic) demonstrates an improvement of up to 6.5% over state-of-the-art GAD methods. The code is available at: https://github.com/karish-grover/curvgad.

How to best develop foundational models for time series forecasting remains an important open question. Tokenization is a crucial consideration in this effort: what is an effective discrete vocabulary for a real-valued sequential input? To address this question, we develop WaveToken, a wavelet-based tokenizer that allows models to learn complex representations directly in the space of time-localized frequencies. Our method first scales and decomposes the input time series, then thresholds and quantizes the wavelet coefficients, and finally pre-trains an autoregressive model to forecast coefficients for the forecast horizon. By decomposing coarse and fine structures in the inputs, wavelets provide an eloquent and compact language for time series forecasting that simplifies learning. Empirical results on a comprehensive benchmark, including 42 datasets for both in-domain and zero-shot settings, show that WaveToken: i) performs on par or better than recently proposed foundation models for forecasting while using a much smaller vocabulary (1024 tokens), and is competitive with modern deep learning models trained specifically on each dataset; ii) exhibits superior generalization capabilities, achieving the best average rank across all datasets for three complementary metrics; and iii) easily captures complex temporal patterns of practical relevance that are challenging for other recent pre-trained models, including trends, sparse spikes, and non-stationary time series with varying frequencies evolving over time.

Since the seminal work of TabPFN, research on tabular foundation models (TFMs) based on in-context learning (ICL) has challenged long-standing paradigms in machine learning. Without seeing any real-world data, models pretrained on purely synthetic datasets generalize remarkably well across diverse datasets, often using only a moderate number of in-context examples. This shifts the focus in tabular machine learning from model architecture design to the design of synthetic datasets, or, more precisely, to the prior distributions that generate them. Yet the guiding principles for prior design remain poorly understood. This work marks the first attempt to address the gap. We systematically investigate and identify key properties of synthetic priors that allow pretrained TFMs to generalize well. Based on these insights, we introduce Mitra, a TFM trained on a curated mixture of synthetic priors selected for their diversity, distinctiveness, and performance on real-world tabular data. Mitra consistently outperforms state-of-the-art TFMs, such as TabPFNv2 and TabICL, across both classification and regression benchmarks, with better sample efficiency.

Cybersecurity analysts face an escalating threat landscape where zero-day malware families emerge daily, requiring automated detection systems that can identify previously unseen threats while maintaining low false positive rates in operational environments. Existing malware detection systems excel at classifying known families but struggle with novel variants, often treating novelty detection as a separate post-processing step that leads to excessive false alarms and missed threats.We present CentroidEmbed, a unified framework that integrates prototype learning directly within graph neural networks for robust malware family detection and zero-day threat identification. Our approach addresses critical operational challenges by jointly optimizing for both accurate classification of known malware families and reliable detection of novel threats within a single model architecture. The system introduces three key security-focused innovations: (1) learnable malware family centroids that adapt to evolving threat patterns through gradient-based refinement, (2) a specialized cosine-based triplet center loss that creates distinct behavioral signatures for each family while maintaining separation from potential novel threats, and (3) a parallel novelty detection component that identifies suspicious samples without requiring prior knowledge of specific attack vectors.Evaluated on real-world malware datasets using temporally realistic chronological splits that simulate operational deployment scenarios, CentroidEmbed demonstrates substantial improvements over existing security-focused detection methods. On the BODMAS corpus containing over 57,000 malware samples across 500+ families, our approach achieves 77.2% precision and 55.6% recall for novel family detection, significantly reducing false alarms while maintaining high detection rates for zero-day threats. These results demonstrate that integrating prototype learning within neural architectures creates more operationally viable detection systems compared to conventional classification approaches with post-hoc anomaly detection, providing security analysts with a practical tool for identifying emerging threats in dynamic cyber environments.

Do real-world hypergraphs obey any patterns? Are power laws fundamental in hypergraphs as they are in real-world graphs? What generator can reproduce these patterns? A hypergraph is a generalization of a conventional graph, and it consists of nodes and hyperedges, with each hyperedge joining any number of nodes. Hypergraphs are adept at representing group interactions where two or more entities interact simultaneously, such as collaborative research and group discussions. In a wide range of real-world hypergraphs, we discover power-law or log-logistic distributions in eight structural properties. To simulate these observed patterns, we introduce HyRec, a tractable and realistic generative model leveraging the Kronecker product. We mathematically demonstrate that HyRec accurately reproduces both the patterns we observed and typical evolutionary trends found in real-world hypergraphs. To fit the parameters of HyRec to large-scale hypergraphs, we design SingFit, a fast and space-efficient algorithm successfully applied to eleven real-world hypergraphs with up to one million nodes and hyperedges. This paper makes the following contributions: (a) Discoveries: we identify multiple patterns that real-world hypergraphs obey, (b) Model: we propose HyRec, a tractable and realistic model capable of reproducing real-world hypergraphs efficiently (spec., with fewer than 1,000 parameters) with the support of SingFit, and (c) Proofs: we prove that HyRec adheres to these patterns.

We propose OmniMatch , a novel joinability discovery technique, specifically tailored for the needs of data products : cohesive curated collections of tabular datasets. OmniMatch combines multiple column-pair similarity measures leveraging self-supervised Graph Neural Networks (GNNs). OmniMatch 's GNN captures column relatedness by leveraging graph neighborhood information, significantly improving the recall of joinability discovery tasks. At the same time, OmniMatch increases its precision by augmenting its training data with negative column join examples through an automated negative example generation process. Compared to the state-of-the-art, OmniMatch exhibits up to 14% higher effectiveness in F1 score and AUC without relying on individual, user-provided thresholds for each similarity metric.

Honeybees, as natural crop pollinators, play a significant role in biodiversity and food production for human civilization. Bees actively regulate hive temperature (homeostasis) to maintain a colony’s proper functionality. Deviations from usual thermoregulation behavior due to external stressors (e.g., extreme environmental temperature, parasites, pesticide exposure) indicate an impending colony collapse. Anticipating such threats by forecasting hive temperature and finding changes in temperature patterns would allow beekeepers to take early preventive measures and avoid critical issues. In that case, how can we model bees’ thermoregulation behavior for an interpretable and effective hive monitoring system? In this article, we propose the principled Electronic Bee-Veterinarian Plus (EBV+) method based on the thermal diffusion equation and a novel “ sigmoid ” feedback-loop (P) controller for analyzing hive health with the following properties: (i) it is effective on multiple, real-world beehive time sequences (recorded and streaming), (ii) it is explainable with only a few parameters (e.g., hive health factor) that beekeepers can easily quantify and trust, (iii) it issues proactive alerts to beekeepers before any potential issue affecting homeostasis becomes detrimental, and (iv) it is scalable with a time complexity of \(O(t)\) for reconstructing and \(O(t\times m)\) for finding m cuts of a sequence with t time-ticks. Experimental results on multiple real-world time sequences showcase the potential and practical feasibility of EBV+. Our method yields accurate forecasting (up to 72% improvement in RMSE) with up to 600 times fewer parameters compared to baselines (ARX, seasonal ARX, Holt-winters, and DeepAR), as well as detects discontinuities and raises alerts that coincide with domain experts’ opinions. Moreover, EBV+ is scalable and fast, taking less than 1 minute on a stock laptop to reconstruct 2 months of sensor data.

Human trafficking (HT) for forced sexual exploitation, often described as modern-day slavery, is a pervasive problem that affects millions of people worldwide. Perpetrators of this crime post advertisements (ads) on behalf of their victims on adult service websites (ASW). These websites typically contain hundreds of thousands of ads including those posted by independent escorts, massage parlor agencies and spammers (fake ads). Detecting suspicious activity in these ads is difficult and developing data-driven methods is challenging due to the hard-to-label, complex and sensitive nature of the data. In this paper, we propose T-Net, which unlike previous solutions, formulates this problem as weakly supervised classification. Since it takes several months to years to investigate a case and obtain a single definitive label, we design domain-specific signals or indicators that provide weak labels. T-Net also looks into connections between ads and models the problem as a graph learning task instead of classifying ads independently. We show that T-Net outperforms all baselines on a real-world dataset of ads by 7% average weighted F1 score. Given that this data contains personally identifiable information, we also present a realistic data generator and provide the first publicly available dataset in this domain which may be leveraged by the wider research community.

This paper introduces OrthoReg, a simple yet effective Graph-regularized MLP model for semi-supervised node representation learning. We first demonstrate, through empirical observations and theoretical analysis, that node embeddings learned from conventional GR-MLPs suffer from the over-correlation issue. This issue arises when a few dominant singular values overwhelm the embedding space, leading to the limited expressive power of the learned node representations. To mitigate this problem, we propose a novel GR-MLP model called OrthoReg. By incorporating a soft regularization loss on the correlation matrix of node embeddings, OrthoReg explicitly encourages orthogonal node representations, effectively avoiding over-correlated representations. Compared to the currently popular GNN models, our OrthoReg possesses two distinct advantages: 1) Much faster inference speed, particularly for large-scale graphs. 2) Significantly superior performance in inductive cold-start settings. Experiments on semi-supervised node classification tasks, together with the extensive ablation studies, have demonstrated the effectiveness of the proposed designs.

Costas Mavromatis, Balasubramaniam Srinivasan, Zhengyuan Shen, Jiani Zhang, Huzefa Rangwala, Christos Faloutsos, George Karypis. Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing. 2024.

How can we effectively generate missing data transformations among tables in a data repository? Multiple versions of the same tables are generated from the iterative process when data scientists and machine learning engineers fine-tune their ML pipelines, making incremental improvements. This process often involves data transformation and augmentation that produces an augmented table based on its base version and related tables. However, data transformations are often not well-documented or completely missing, resulting in poor traceability, reproducibility and explainability of ML pipelines. In this paper, we propose DATALoRE, a framework that explains data changes between an initial dataset and its augmented version to improves traceability. Given a base table, DATALoRE first discovers its potentially related tables from the data repository using a variety of data discovery techniques. DATALoRE then effectively leverages a large language model (LLM) to generate a variety of data transformations that lead to the augmented table. DATALoRE validates these transformations and selects the minimum number of related tables to ensure traceability and reproducibility of the ML pipelines. A preliminary experiment shows that DATALoRE is able to effectively recovery data transformations on two benchmark datasets.

How could we have an outlier detector that works even with nondimensional data, and ranks together both singleton microclusters (‘one-off’ outliers) and nonsingleton microclusters by their anomaly scores? How to obtain scores that are prin-cipled in one scalable and ‘hands-off’ manner? Microclusters of outliers indicate coalition or repetition in fraud activities, etc.; their identification is thus highly desirable. This paper presents Mccatch: a new algorithm that detects microclusters by leveraging our proposed ‘Oracle’ plot (1NN Distance versus Group 1NN Distance). We study 31 real and synthetic datasets with up to 1M data elements to show that McCatchi's the only method that answers both of the questions above; and, it outperforms 11 other methods, especially when the data has non-singleton microclusters or is nondimensional. We also showcase McCATCH'S ability to detect meaningful microclusters in graphs, fingerprints, logs of network connections, text data, and satellite imagery. For example, it found a 30-elements microcluster of confirmed ‘Denial of Service’ attacks in the network logs, taking only ~3 minutes for 222K data elements on a stock desktop.

Large Language Models (LLMs) trained on large volumes of data excel at various natural language tasks, but they cannot handle tasks requiring knowledge that has not been trained on previously. One solution is to use a retriever that fetches relevant information to expand LLM's knowledge scope. However, existing textual-oriented retrieval-based LLMs are not ideal on structured table data due to diversified data modalities and large table sizes. In this work, we propose OpenTab, an open-domain table reasoning framework powered by LLMs. Overall, OpenTab leverages table retriever to fetch relevant tables and then generates SQL programs to parse the retrieved tables efficiently. Utilizing the intermediate data derived from the SQL executions, it conducts grounded inference to produce accurate response. Extensive experimental evaluation shows that OpenTab significantly outperforms baselines in both open- and closed-domain settings, achieving up to 21.5% higher accuracy. We further run ablation studies to validate the efficacy of our proposed designs of the system.

Given a node-attributed graph, and a graph task (link prediction or node classification), can we tell if a graph neural network (GNN) will perform well? More specifically, do the graph structure and the node features carry enough usable information for the task? Our goals are (1) to develop a fast tool to measure how much information is in the graph structure and in the node features, and (2) to exploit the information to solve the task, if there is enough. We propose NetInfoF, a framework including NetInfoF_Probe and NetInfoF_Act, for the measurement and the exploitation of network usable information (NUI), respectively. Given a graph data, NetInfoF_Probe measures NUI without any model training, and NetInfoF_Act solves link prediction and node classification, while two modules share the same backbone. In summary, NetInfoF has following notable advantages: (a) General, handling both link prediction and node classification; (b) Principled, with theoretical guarantee and closed-form solution; (c) Effective, thanks to the proposed adjustment to node similarity; (d) Scalable, scaling linearly with the input size. In our carefully designed synthetic datasets, NetInfoF correctly identifies the ground truth of NUI and is the only method being robust to all graph scenarios. Applied on real-world datasets, NetInfoF wins in 11 out of 12 times on link prediction compared to general GNN baselines.

Recent advances in tabular data generation have greatly enhanced synthetic data quality. However, extending diffusion models to tabular data is challenging due to the intricately varied distributions and a blend of data types of tabular data. This paper introduces TabSyn, a methodology that synthesizes tabular data by leveraging a diffusion model within a variational autoencoder (VAE) crafted latent space. The key advantages of the proposed Tabsyn include (1) Generality: the ability to handle a broad spectrum of data types by converting them into a single unified space and explicitly capturing inter-column relations; (2) Quality: optimizing the distribution of latent embeddings to enhance the subsequent training of diffusion models, which helps generate high-quality synthetic data; (3) Speed: much fewer number of reverse steps and faster synthesis speed than existing diffusion-based methods. Extensive experiments on six datasets with five metrics demonstrate that Tabsyn outperforms existing methods. Specifically, it reduces the error rates by 86% and 67% for column-wise distribution and pair-wise column correlation estimations compared with the most competitive baselines. The code has been made available at https://github.com/amazon-science/tabsyn.

While graph machine learning, and notably graph neural networks (GNNs), have gained immense traction in recent years, application is predicated on access to a known input graph upon which predictive models can be trained. And indeed, within the most widely-studied public evaluation benchmarks such graphs are provided, with performance comparisons conditioned on curated data explicitly adhering to this graph. However, in real-world industrial applications, the situation is often quite different. Instead of a known graph, data are originally collected and stored across multiple tables in a repository, at times with ambiguous or incomplete relational structure. As such, to leverage the latest GNN architectures it is then up to a skilled data scientist to first manually construct a graph using intuition and domain knowledge, a laborious process that may discourage adoption in the first place. To narrow this gap and broaden the applicability of graph ML, we survey existing tools and strategies that can be combined to address the more fundamental problem of predictive tabular modeling over data native to multiple tables, with no explicit relational structure assumed a priori. This involves tracing a comprehensive path through related table join discovery and fuzzy table joining, column alignment, automated relational database (RDB) construction, extracting graphs from RDBs, graph sampling, and finally, graph-centric trainable predictive architectures. Although efforts to build deployable systems that integrate all of these components while minimizing manual effort remain in their infancy, this survey will nonetheless reduce barriers to entry and help steer the graph ML community towards promising research directions and wider real-world impact.

Graph machine learning (GML) is effective in many business applications. However, making GML easy to use and applicable to industry applications with massive datasets remain challenging. We developed GraphStorm, which provides an end-to-end solution for scalable graph construction, graph model training and inference. GraphStorm has the following desirable properties: (a) Easy to use: it can perform graph construction and model training and inference with just a single command; (b) Expert-friendly: GraphStorm contains many advanced GML modeling techniques to handle complex graph data and improve model performance; (c) Scalable: every component in GraphStorm can operate on graphs with billions of nodes and can scale model training and inference to different hardware without changing any code. GraphStorm has been used and deployed for over a dozen billion-scale industry applications after its release in May 2023. It is open-sourced in Github: https://github.com/awslabs/graphstorm.

Given a relational database (RDB), how can we predict missing column values in some target table of interest? Although RDBs store vast amounts of rich, informative data spread across interconnected tables, the progress of predictive machine learning models as applied to such tasks arguably falls well behind advances in other domains such as computer vision or natural language processing. This deficit stems, at least in part, from the lack of established/public RDB benchmarks as needed for training and evaluation purposes. As a result, related model development thus far often defaults to tabular approaches trained on ubiquitous single-table benchmarks, or on the relational side, graph-based alternatives such as GNNs applied to a completely different set of graph datasets devoid of tabular characteristics. To more precisely target RDBs lying at the nexus of these two complementary regimes, we explore a broad class of baseline models predicated on: (i) converting multi-table datasets into graphs using various strategies equipped with efficient subsampling, while preserving tabular characteristics; and (ii) trainable models with well-matched inductive biases that output predictions based on these input subgraphs. Then, to address the dearth of suitable public benchmarks and reduce siloed comparisons, we assemble a diverse collection of (i) large-scale RDB datasets and (ii) coincident predictive tasks. From a delivery standpoint, we operationalize the above four dimensions (4D) of exploration within a unified, scalable open-source toolbox called 4DBInfer; please see https://github.com/awslabs/multi-table-benchmark .

Honeybees are vital for pollination and food production. Among many factors, extreme temperature (e.g., due to climate change) is particularly dangerous for bee health. Anticipating such extremities would allow beekeepers to take early preventive action. Thus, given sensor (temperature) time series data from beehives, how can we find patterns and do forecasting? Forecasting is crucial as it helps spot unexpected behavior and thus issue warnings to the beekeepers. In that case, what are the right models for forecasting? ARIMA, RNNs, or something else?

How can we enhance the node features acquired from Pretrained Models (PMs) to better suit downstream graph learning tasks? Graph Neural Networks (GNNs) have become the state-of-the-art approach for many high-impact, real-world graph applications. For feature-rich graphs, a prevalent practice involves directly utilizing a PM to generate features. Nevertheless, this practice is suboptimal as the node features extracted from PMs are graph-agnostic and prevent GNNs from fully utilizing the potential correlations between the graph structure and node features, leading to a decline in GNN performance. In this work, we seek to improve the node features obtained from a PM for graph tasks and introduce TouchUp-G, a "Detect & Correct" approach for refining node features extracted from PMs. TouchUp-G detects the alignment using a novel feature homophily metric and corrects the misalignment through a simple touchup on the PM. It is (a) General: applicable to any downstream graph task; (b) Multi-modal: able to improve raw features of any modality; (c) Principled: it is closely related to a novel metric, feature homophily, which we propose to quantify the alignment between the graph structure and node features; (d) Effective: achieving state-of-the-art results on four real-world datasets spanning different tasks and modalities.

Predicting how social networks change in the future is important in many applications. Results in social network research have shown that the change in the network can be explained by a small number of concepts, such as "homophily" and "transitivity". However, existing prediction methods require many latent features that are not connected to such concepts, making the methods' black boxes and their prediction results difficult to interpret, making them harder to derive scientific knowledge about social networks. In this study, we propose NetEvolve a novel multi-agent reinforcement learning-based method that predicts changes in a given social network. Given a sequence of changes as training data, NetEvolve learns the characteristics of the nodes with interpretable features, such as how the node feels rewards for connecting with similar people and the cost of the connection itself. Based on the learned feature, NetEvolve makes a forecast based on multi-agent simulation. The method achieves comparable or better accuracy than existing methods in predicting network changes in real-world social networks while keeping the prediction results interpretable.

Forecasting extrapolates the values of a time series into the future, and is crucial to optimize core operations for many businesses and organizations. Building machine learning (ML)-based forecasting applications presents a challenge though, due to non-stationary data and large numbers of time series. As there is no single dominating approach to forecasting, forecasting systems have to support a wide variety of approaches, ranging from deep learning-based methods to classical methods built on probabilistic modelling. We revisit our earlier work on a monolithic platform for forecasting from VLDB 2017, and describe how we evolved it into a modern forecasting stack consisting of several layers that support a wide range of forecasting needs and automate common tasks like model selection. This stack leverages our open source forecasting libraries GluonTS and AutoGluon-TimeSeries , the scalable ML platform SageMaker , and forms the basis of the no-code forecasting solutions ( SageMaker Canvas and Amazon Forecast ), available in the Amazon Web Services cloud. We give insights into the predictive performance of our stack and discuss learnings from using it to provision resources for the cloud database services DynamoDB, Redshift and Athena.

Given a million-scale dataset of who-calls-whom data containing imperfect labels, how can we detect existing and new fraud patterns? We propose TgrApp, which extracts carefully designed features and provides visualizations to assist analysts in spotting fraudsters and suspicious behavior. Our TgrApp method has the following properties: (a) Scalable, as it is linear on the input size; and (b) Effective, as it allows natural interaction with human analysts, and is applicable in both supervised and unsupervised settings.

Given a large, unlabeled set of Electronic Health Records (EHRs) acquired from multiple hospitals, how can we analyze the available entities and identify relationships in the data? Also, how can we perform Exploratory Data Analysis (EDA) over such EHR data? Many medical institutions generate EHRs as tabular data with entities and attributes in common. However, due to a large number of records, attributes, and high cardinality, exploring the different datasets and finding patterns and insights become laborious and prone to errors. In this work, we propose GraF- Eda for EDA over EHR data from different institutions. GraF-EDA models EHRs as time-evolving graphs, allowing the interoperability of such data into a single representation. We extract meaningful features from the graph nodes and provide intuitive visualizations to improve data explainability. We evaluate GraF-EDA with four COVID-19 datasets from hospitals of the São Paulo state, Brazil, resulting in million-scale graphs. Our method identified correlations, similarities and dissimilarities among medical treatments, exams, clinics, and outcomes. With the visual tools provided by GraF-EDA, we were able to spot cases of interest and check more details about them. Our results indicate that GraF-EDA is a fast, effective, open-sourced tool for EDA of EHRs from multiple institutions.

Given a million-scale dataset of who-calls-whom data containing imperfect labels, how can we detect existing and new fraud patterns? We propose CallMine, with carefully designed features and visualizations. Our CallMine method has the following properties: (a) Scalable, being linear on the input size, handling about 35 million records in around one hour on a stock laptop; (b) Effective, allowing natural interaction with human analysts; (c) Flexible, being applicable in both supervised and unsupervised settings; (d) Automatic, requiring no user-defined parameters.

How can we solve semi-supervised node classification in various graphs possibly with noisy features and structures? Graph neural networks (GNNs) have succeeded in many graph mining tasks, but their generalizability to various graph scenarios is limited due to the difficulty of training, hyperparameter tuning, and the selection of a model itself. Einstein said that we should "make everything as simple as possible, but not simpler." We rephrase it into the careful simplicity principle: a carefully-designed simple model can surpass sophisticated ones in real-world graphs. Based on the principle, we propose SlimG for semi-supervised node classification, which exhibits four desirable properties: It is (a) accurate, winning or tying on 10 out of 13 real-world datasets; (b) robust, being the only one that handles all scenarios of graph data (homophily, heterophily, random structure, noisy features, etc.); (c) fast and scalable, showing up to 18 times faster training in million-scale graphs; and (d) interpretable, thanks to the linearity and sparsity. We explain the success of SlimG through a systematic study of the designs of existing GNNs, sanity checks, and comprehensive ablation studies.

The choice of a graph learning (GL) model (i.e., a GL algorithm and its hyperparameter settings) has a significant impact on the performance of downstream tasks. However, selecting the right GL model becomes increasingly difficult and time consuming as more and more GL models are developed. Accordingly, it is of great significance and practical value to equip users of GL with the ability to perform a near-instantaneous selection of an effective GL model without manual intervention. Despite the recent attempts to tackle this important problem, there has been no comprehensive benchmark environment to evaluate the performance of GL model selection methods. To bridge this gap, we present GLEMOS in this work, a comprehensive benchmark for instantaneous GL model selection that makes the following contributions. (i) GLEMOS provides extensive benchmark data for fundamental GL tasks, i.e., link prediction and node classification, including the performances of 366 models on 457 graphs on these tasks. (ii) GLEMOS designs multiple evaluation settings, and assesses how effectively representative model selection techniques perform in these different settings. (iii) GLEMOS is designed to be easily extended with new models, new graphs, and new performance records. (iv) Based on the experimental results, we discuss the limitations of existing approaches and highlight future research directions. To promote research on this significant problem, we make the benchmark data and code publicly available at https://namyongpark.github.io/glemos.

Transparency and accountability have become major concerns for black-box machine learning (ML) models. Proper explanations for the model behavior increase model transparency and help researchers develop more accountable models. Graph neural networks (GNN) have recently shown superior performance in many graph ML problems than traditional methods, and explaining them has attracted increased interest. However, GNN explanation for link prediction (LP) is lacking in the literature. LP is an essential GNN task and corresponds to web applications like recommendation and sponsored search on web. Given existing GNN explanation methods only address node/graph-level tasks, we propose Path-based GNN Explanation for heterogeneous Link prediction (PaGE-Link) that generates explanations with connection interpretability, enjoys model scalability, and handles graph heterogeneity. Qualitatively, PaGE-Link can generate explanations as paths connecting a node pair, which naturally captures connections between the two nodes and easily transfer to human-interpretable explanations. Quantitatively, explanations generated by PaGE-Link improve AUC for recommendation on citation and user-item graphs by 9 - 35% and are chosen as better by 78.79% of responses in human evaluation.

MOTIVATION: Many real-world problems can be modeled as annotated graphs. Scalable graph algorithms that extract actionable information from such data are in demand since these graphs are large, varying in topology, and have diverse node/edge annotations. When these graphs change over time they create dynamic graphs, and open the possibility to find patterns across different time points. In this article, we introduce a scalable algorithm that finds unique dense regions across time points in dynamic graphs. Such algorithms have applications in many different areas, including the biological, financial, and social domains. RESULTS: There are three important contributions to this manuscript. First, we designed a scalable algorithm, USNAP, to effectively identify dense subgraphs that are unique to a time stamp given a dynamic graph. Importantly, USNAP provides a lower bound of the density measure in each step of the greedy algorithm. Second, insights and understanding obtained from validating USNAP on real data show its effectiveness. While USNAP is domain independent, we applied it to four non-small cell lung cancer gene expression datasets. Stages in non-small cell lung cancer were modeled as dynamic graphs, and input to USNAP. Pathway enrichment analyses and comprehensive interpretations from literature show that USNAP identified biologically relevant mechanisms for different stages of cancer progression. Third, USNAP is scalable, and has a time complexity of O(m+mc log nc+nc log nc), where m is the number of edges, and n is the number of vertices in the dynamic graph; mc is the number of edges, and nc is the number of vertices in the collapsed graph. AVAILABILITY AND IMPLEMENTATION: The code of USNAP is available at https://www.cs.utoronto.ca/~juris/data/USNAP22.

Similarity search is a core analytical task, and its performance critically depends on the choice of distance measure. For time-series querying, elastic measures achieve state-of-the-art accuracy but are computationally expensive. Thus, fast lower bounding (LB) measures prune unnecessary comparisons with elastic distances to accelerate similarity search. Despite decades of attention, there has never been a study to assess the progress in this area. In addition, the research has disproportionately focused on one popular elastic measure, while other accurate measures have received little or no attention. Therefore, there is merit in developing a framework to accumulate knowledge from previously developed LBs and eliminate the notoriously challenging task of designing separate LBs for each elastic measure. In this paper, we perform the first comprehensive study of 11 LBs spanning 5 elastic measures using 128 datasets. We identify four properties that constitute the effectiveness of LBs and propose the Generalized Lower Bounding (GLB) framework to satisfy all desirable properties. GLB creates cache-friendly summaries, adaptively exploits summaries of both query and target time series, and captures boundary distances in an unsupervised manner. GLB outperformsallLBs in speedup (e.g., up to 13.5× faster against the strongest LB in terms of pruning power), establishes new state-of-the-art results for the 5 elastic measures, and provides the first LBs for 2 elastic measures with no known LBs. Overall, GLB enables the effective development of LBs to facilitate fast similarity search.

Given a million escort advertisements, how can we spot near-duplicates? Such micro-clusters of ads are usually signals of human trafficking (HT). How can we summarize them to convince law enforcement to act? Spotting micro-clusters of near-duplicate documents is useful in multiple, additional settings, including spam-bot detection in Twitter ads, plagiarism, and more. We present InfoShield , which makes the following contributions: practical , being scalable and effective on real data; parameter-free and principled , requiring no user-defined parameters; interpretable , finding a document to be the cluster representative, highlighting all the common phrases, and automatically detecting “slots” (i.e., phrases that differ in every document); and generalizable , beating or matching domain-specific methods in Twitter bot detection and HT detection, respectively, as well as being language independent. Interpretability is particularly important for the anti-HT domain, where law enforcement must visually inspect ads. Our experiments on real data show that InfoShield correctly identifies Twitter bots with an F1 score over 90% and detects HT ads with 84% precision. Moreover, it is scalable, requiring about 8 hours for 4 million documents on a stock laptop. Our incremental version, DeltaShield , allows for fast, incremental updates, with minor loss of accuracy.

Law enforcement and domain experts can detect human trafficking (HT) in online escort websites by analyzing suspicious clusters of connected ads. How can we explain clustering results intuitively and interactively, visualizing potential evidence for experts to analyze? We present TRAFFICVIS, the first interface for cluster-level HT detection and labeling. Developed through months of participatory design with domain experts, TRAFFICVIS provides coordinated views in conjunction with carefully chosen backend algorithms to effectively show spatio-temporal and text patterns to a wide variety of anti-HT stakeholders. We build upon state-of-the-art text clustering algorithms by incorporating shared metadata as a signal of connected and possibly suspicious activity, then visualize the results. Domain experts can use TRAFFICVIS to label clusters as HT, or other, suspicious, but non-HT activity such as spam and scam, quickly creating labeled datasets to enable further HT research. Through domain expert feedback and a usage scenario, we demonstrate TRAFFICVIS's efficacy. The feedback was overwhelmingly positive, with repeated high praises for the usability and explainability of our tool, the latter being vital for indicting possible criminals.

Given a large, time-evolving graph of who-calls-whom-when, how can we help analysts find anomalies and fraudsters? How can we explain our decisions? We provide TgraphSpot, which carefully extracts features that are often related to fraud; and which provides informative, interactive plots that help analysts zoom down to the few strange nodes. We present the architecture and design decisions of TgraphSpot. Thanks to our careful feature-extraction algorithms, it scales linearly, taking 2.5 hours on a stock laptop, to process 29 million phone calls. More importantly, when applied on a real dataset of millions of phone calls, it discovered suspicious nodes; experts confirmed that those nodes are fraudsters that had been undetected so far.

Fraud mechanisms have evolved from isolated actions performed by single individuals to complex criminal networks. This paper aims to contribute to the identification of potentially relevant nodes in fraud networks. Whilst traditional methods for fraud detection rely on identifying abnormal patterns, this paper proposes STARBRIDGE: a new linear and scalable, ranked out, parameter free method to identify fraudulent nodes and rings based on Bridging, Influence and Control metrics. This is applied to the telecommunications domain where fraudulent nodes form a star-bridge-star pattern. Over 75% of nodes involved in fraud denote control, bridging centrality and doubled the influence scores, when compared to non-fraudulent nodes in the same role, stars and bridges being chief positions.

This paper describes a new method for representing embedding tables of graph neural networks (GNNs) more compactly via tensor-train (TT) decomposition. We consider the scenario where (a) the graph data that lack node features, thereby requiring the learning of embeddings during training; and (b) we wish to exploit GPU platforms, where smaller tables are needed to reduce host-to-GPU communication even for large-memory GPUs. The use of TT enables a compact parameterization of the embedding, rendering it small enough to fit entirely on modern GPUs even for massive graphs. When combined with judicious schemes for initialization and hierarchical graph partitioning, this approach can reduce the size of node embedding vectors by 1,659 times to 81,362 times on large publicly available benchmark datasets, achieving comparable or better accuracy and significant speedups on multi-GPU systems. In some cases, our model without explicit node features on input can even match the accuracy of models that use node features.

If a person changes their interests and opinions, how will that change and affect their friendships? Conversely, when a person changes who they are friends with, how will that affect their interests and opinions. Clearly, the person's relationships and interests are related. For example, if several of a person's friends are smokers, but they themselves are not, they are likely to either start smoking as well or to reduce their relationships with the smokers and make new, non-smoking friends. We propose DualCast, a method for predicting the evolution of friendship edges between nodes, as well as the attribute values (which represent opinions and preferences), of nodes in an attributed network. One of the main contributions of the present study is the ability to assume and estimate two scores for each node: the influence (its power to influence neighbors) and its susceptibility (how easily it can be influenced). Our DualCast has the following novel benefits: (A) Expressive: it can capture when links between nodes are dropped, as well as polarization that occurs with changes in interests, (B) Scalable: its performance is linear with input size, (C) Accurate: it is up to 8% more accurate in forecasting links between nodes, and up to 20% more accurate for attribute-values, when tested on publicly available, real datasets with 100K nodes.

Sequential recommendation aims to model dynamic user behavior from historical interactions. Self-attentive methods have proven effective at capturing short-term dynamics and long-term preferences. Despite their success, these approaches still struggle to model sparse data, on which they struggle to learn high-quality item representations. We propose to model user dynamics from shopping intents and interacted items simultaneously. The learned intents are coarse-grained and work as prior knowledge for item recommendation. To this end, we present a coarse-to-fine self-attention framework, namely CaFe, which explicitly learns coarse-grained and fine-grained sequential dynamics. Specifically, CaFe first learns intents from coarse-grained sequences which are dense and hence provide high-quality user intent representations. Then, CaFe fuses intent representations into item encoder outputs to obtain improved item representations. Finally, we infer recommended items based on representations of items and corresponding intents. Experiments on sparse datasets show that CaFe outperforms state-of-the-art self-attentive recommenders by 44.03% [email protected] on average.

How can we perform knowledge reasoning over temporal knowledge graphs (TKGs)? TKGs represent facts about entities and their relations, where each fact is associated with a timestamp. Reasoning over TKGs, i.e., inferring new facts from time-evolving KGs, is crucial for many applications to provide intelligent services. However, despite the prevalence of real-world data that can be represented as TKGs, most methods focus on reasoning over static knowledge graphs, or cannot predict future events. In this paper, we present a problem formulation that unifies the two major problems that need to be addressed for an effective reasoning over TKGs, namely, modeling the event time and the evolving network structure. Our proposed method EvoKG jointly models both tasks in an effective framework, which captures the ever-changing structural and temporal dynamics in TKGs via recurrent event modeling, and models the interactions between entities based on the temporal neighborhood aggregation framework. Further, EvoKG achieves an accurate modeling of event time, using flexible and efficient mechanisms based on neural density estimation. Experiments show that EvoKG outperforms existing methods in terms of effectiveness (up to 77% and 116% more accurate time and link prediction) and efficiency.

Given entities and their interactions in the web data, which may have occurred at different time, how can we find communities of entities and track their evolution? In this paper, we approach this important task from graph clustering perspective. Recently, state-of-the-art clustering performance in various domains has been achieved by deep clustering methods. Especially, deep graph clustering (DGC) methods have successfully extended deep clustering to graph-structured data by learning node representations and cluster assignments in a joint optimization framework. Despite some differences in modeling choices (e.g., encoder architectures), existing DGC methods are mainly based on autoencoders and use the same clustering objective with relatively minor adaptations. Also, while many real-world graphs are dynamic, previous DGC methods considered only static graphs. In this work, we develop CGC, a novel end-to-end framework for graph clustering, which fundamentally differs from existing methods. CGC learns node embeddings and cluster assignments in a contrastive graph learning framework, where positive and negative samples are carefully selected in a multi-level scheme such that they reflect hierarchical community structures and network homophily. Also, we extend CGC for time-evolving data, where temporal graph clustering is performed in an incremental learning fashion, with the ability to detect change points. Extensive evaluation on real-world graphs demonstrates that the proposed CGC consistently outperforms existing methods.

Automatic extraction of product attributes from their textual descriptions is essential for online shopper experience. One inherent challenge of this task is the emerging nature of e-commerce products — we see new types of products with their unique set of new attributes constantly. Most prior works on this matter mine new values for a set of known attributes but cannot handle new attributes that arose from constantly changing data. In this work, we study the attribute mining problem in an open-world setting to extract novel attributes and their values. Instead of providing comprehensive training data, the user only needs to provide a few examples for a few known attribute types as weak supervision. We propose a principled framework that first generates attribute value candidates and then groups them into clusters of attributes. The candidate generation step probes a pre-trained language model to extract phrases from product titles. Then, an attribute-aware fine-tuning method optimizes a multitask objective and shapes the language model representation to be attribute-discriminative. Finally, we discover new attributes and values through the self-ensemble of our framework, which handles the open-world challenge. We run extensive experiments on a large distantly annotated development set and a gold standard human-annotated test set that we collected. Our model significantly outperforms strong baselines and can generalize to unseen attributes and product types.

Given a stream of graph edges from a dynamic graph, how can we assign anomaly scores to edges in an online manner, for the purpose of detecting unusual behavior, using constant time and memory? Existing approaches aim to detect individually surprising edges. In this work, we propose Midas , which focuses on detecting microcluster anomalies , or suddenly arriving groups of suspiciously similar edges, such as lockstep behavior, including denial of service attacks in network traffic data. We further propose Midas -F, to solve the problem by which anomalies are incorporated into the algorithm’s internal states, creating a “poisoning” effect that can allow future anomalies to slip through undetected. Midas -F introduces two modifications: (1) we modify the anomaly scoring function, aiming to reduce the “poisoning” effect of newly arriving edges; (2) we introduce a conditional merge step, which updates the algorithm’s data structures after each time tick, but only if the anomaly score is below a threshold value, also to reduce the “poisoning” effect. Experiments show that Midas -F has significantly higher accuracy than Midas . In general, the algorithms proposed in this work have the following properties: (a) they detects microcluster anomalies while providing theoretical guarantees about the false positive probability; (b) they are online, thus processing each edge in constant time and constant memory, and also processes the data orders-of-magnitude faster than state-of-the-art approaches; and (c) they provides up to 62% higher area under the receiver operating characteristic curve than state-of-the-art approaches.

In a cloud of m-dimensional data points, how would we spot, as well as rank, both single-point- as well as group-anomalies? We are the first to generalize anomaly detection in two dimensions: The first dimension is that we handle both point-anomalies, as well as group-anomalies, under a unified view - we shall refer to them as generalized anomalies. The second dimension is that Gen2Out not only detects, but also ranks, anomalies in suspiciousness order. Detection, and ranking, of anomalies has numerous applications: For example, in EEG recordings of an epileptic patient, an anomaly may indicate a seizure; in computer network traffic data, it may signify a power failure, or a DoS/DDoS attack.We start by setting some reasonable axioms; surprisingly, none of the earlier methods pass all the axioms. Our main contribution is the Gen2Out algorithm, that has the following desirable properties: (a) Principled and Sound anomaly scoring that obeys the axioms for detectors, (b) Doubly-general in that it detects, as well as ranks generalized anomaly– both point- and group-anomalies, (c) Scalable, it is fast and scalable, linear on input size. (d) Effective, experiments on real-world epileptic recordings (200GB) demonstrate effectiveness of Gen2Out as confirmed by clinicians. Experiments on 27 real-world benchmark datasets show that Gen2Out detects ground truth groups, matches or outperforms point-anomaly baseline algorithms on accuracy, with no competition for group-anomalies and requires about 2 minutes for 1 million data points on a stock machine.

Given a million escort advertisements, how can we spot near-duplicates? Such micro-clusters of ads are usually signals of human trafficking. How can we summarize them, visually, to convince law enforcement to act? Can we build a general tool that works for different languages? Spotting micro-clusters of near-duplicate documents is useful in multiple, additional settings, including spam-bot detection in Twitter ads, plagiarism, and more.We present INFOSHIELD, which makes the following contributions: (a) Practical, being scalable and effective on real data, (b) Parameter-free and Principled, requiring no user-defined parameters, (c) Interpretable, finding a document to be the cluster representative, highlighting all the common phrases, and automatically detecting "slots", i.e. phrases that differ in every document; and (d) Generalizable, beating or matching domain-specific methods in Twitter bot detection and human trafficking detection respectively, as well as being language-independent finding clusters in Spanish, Italian, and Japanese. Interpretability is particularly important for the anti human-trafficking domain, where law enforcement must visually inspect ads.Our experiments on real data show that INFOSHIELD correctly identifies Twitter bots with an F1 score over 90% and detects human-trafficking ads with 84% precision. Moreover, it is scalable, requiring about 8 hours for 4 million documents on a stock laptop.

Existing recommender systems in the e-commerce domain primarily focus on generating a set of relevant items as recommendations; however, few existing systems utilize underlying item attributes as a key organizing principle in presenting recommendations to users. Mining important attributes of items from customer perspectives and presenting them along with item sets as recommendations can provide users more explainability and help them make better purchase decision. In this work, we generalize the attribute-aware item-set recommendation problem, and develop a new approach to generate sets of items (recommendations) with corresponding important attributes (explanations) that can best justify why the items are recommended to users. In particular, we propose a system that learns important attributes from historical user behavior to derive item set recommendations, so that an organized view of recommendations and their attribute-driven explanations can help users more easily understand how the recommendations relate to their preferences. Our approach is geared towards real world scenarios: we expect a solution to be scalable to billions of items, and be able to learn item and attribute relevance automatically from user behavior without human annotations. To this end, we propose a multi-step learning-based framework called Extract-Expect-Explain (EX3), which is able to adaptively select recommended items and important attributes for users. We experiment on a large-scale real-world benchmark and the results show that our model outperforms state-of-the-art baselines by an 11.35% increase on NDCG with adaptive explainability for item set recommendation.

Given a graph stream, how can we estimate the number of triangles in it using multiple machines with limited storage? Specifically, how should edges be processed and sampled across the machines for rapid and accurate estimation? The count of triangles (i.e., cliques of size three) has proven useful in numerous applications, including anomaly detection, community detection, and link recommendation. For triangle counting in large and dynamic graphs, recent work has focused largely on streaming algorithms and distributed algorithms but little on their combinations for “the best of both worlds.” In this work, we propose CoCoS , a fast and accurate distributed streaming algorithm for estimating the counts of global triangles (i.e., all triangles) and local triangles incident to each node. Making one pass over the input stream, CoCoS carefully processes and stores the edges across multiple machines so that the redundant use of computational and storage resources is minimized. Compared to baselines, CoCoS is: (a) accurate: giving up to smaller estimation error; (b) fast : up to faster, scaling linearly with the size of the input stream; and (c) theoretically sound : yielding unbiased estimates.

Given a stream of graph edges from a dynamic graph, how can we assign anomaly scores to edges in an online manner, for the purpose of detecting unusual behavior, using constant time and memory? Existing approaches aim to detect individually surprising edges. In this work, we propose Midas, which focuses on detecting microcluster anomalies, or suddenly arriving groups of suspiciously similar edges, such as lockstep behavior, including denial of service attacks in network traffic data. Midas has the following properties: (a) it detects microcluster anomalies while providing theoretical guarantees about its false positive probability; (b) it is online, thus processing each edge in constant time and constant memory, and also processes the data 108–505 times faster than state-of-the-art approaches; (c) it provides 46%-52% higher accuracy (in terms of AUC) than state-of-the-art approaches.

Suppose you visit an e-commerce site, and see that 50 users each reviewed almost all of the same 500 products several times each: would you get suspicious? Similarly, given a Twitter follow graph, how can we design principled measures for identifying surprisingly dense subgraphs? Dense subgraphs often indicate interesting structure, such as network attacks in network traffic graphs. However, most existing dense subgraph measures either do not model normal variation, or model it using an Erdős-Renyi assumption - but this assumption has been discredited decades ago. What is the right assumption then? We propose a novel application of extreme value theory to the dense subgraph problem, which allows us to propose measures and algorithms which evaluate the surprisingness of a subgraph probabilistically, without requiring restrictive assumptions (e.g. Erdős-Renyi). We then improve the practicality of our approach by incorporating empirical observations about dense subgraph patterns in real graphs, and by proposing a fast pruning-based search algorithm. Our approach (a) provides theoretical guarantees of consistency, (b) scales quasi-linearly, and (c) outperforms baselines in synthetic and ground truth settings.

How can we spot money laundering in large-scale graph-like accounting datasets? How to identify the most suspicious period in a time-evolving accounting graph? What kind of accounts and events should practitioners prioritize under time constraints? To tackle these crucial challenges in accounting and auditing tasks, we propose a flexible system called AutoAudit, which can be valuable for auditors and risk management professionals. To sum up, there are four major advantages of the proposed system: (a) "Smurfing" Detection, spots nearly 100% of injected money laundering transactions automatically in real-world datasets. (b) Attention Routing, attends to the most suspicious part of time-evolving graphs and provides an intuitive interpretation. (c) Insight Discovery, identifies similar month-pair patterns proved by "success stories" and patterns following Power Laws in log-logistic scales. (d) Scalability and Generality, ensures AutoAudit scales linearly and can be easily extended to other real-world graph datasets. Experiments on various real-world datasets illustrate the effectiveness of our method. To facilitate reproducibility and accessibility, we make the code, figure, and results public at https://github.com/mengchillee/AutoAudit.

If one customer buys a tennis racket, what are the best 3 complementary products to purchase together? 3 tennis ball packs, 3 headbands, 3 overgrips, or 1 of each respectively? Complementary product recommendation (CPR), aiming at providing product suggestions that are often bought together to serve a joint demand, forms a pivotal component of e-commerce service, however, existing methods are far from optimal. Given one product, how to recommend its complementary products of different types is the key problem we tackle in this work. We first conduct an extensive analysis to correct the inaccurate assumptions adopted by existing work to show that co-purchased products are not always complementary and further propose a new strategy to generate clean distant supervision labels for CPR modeling. Moreover, to bridge in the gap from existing work that CPR does not only need relevance modeling but also requires diversity to fulfill the whole purchase demand, we develop a deep learning framework, P-Companion to explicitly model both relevance and diversity. More specifically, given one product with its product type, P-Companion first uses an encoder-decoder network to predict multiple complementary product types, then a transfer metric learning network is developed to project the embedding of query product to each predicted complementary product type subspace and further learn the complementary relationship based on the distant supervision labels within each subspace. The whole framework can be trained from end-to-end and robust to cold-start products attributed to a novel pretrained product embedding module named Product2Vec, based on graph attention networks. Extensive offline experiments show that P-Companion outperforms state-of-the-art baselines by 7.1% increase on the [email protected] score with well-controlled diversity. Production-wise, we deploy P-Companion to provide online recommendations for over 200M products at Amazon and observe significant gains on product sales and profit.

Online recommendation is an essential functionality across a variety of services, including e-commerce and video streaming, where items to buy, watch, or read are suggested to users. Justifying recommendations, i.e., explaining why a user might like the recommended item, has been shown to improve user satisfaction and persuasiveness of the recommendation. In this paper, we develop a method for generating post-hoc justifications that can be applied to the output of any recommendation algorithm. Existing post-hoc methods are often limited in providing diverse justifications, as they either use only one of many available types of input data, or rely on the predefined templates. We address these limitations of earlier approaches by developing J-Recs, a method for producing concise and diverse justifications. J-Recs is a recommendation model-agnostic method that generates diverse justifications based on various types of product and user data (e.g., purchase history and product attributes). The challenge of jointly processing multiple types of data is addressed by designing a principled graph-based approach for justification generation. In addition to theoretical analysis, we present an extensive evaluation on synthetic and real-world data. Our results show that J-Recs satisfies desirable properties of justifications, and efficiently produces effective justifications, matching user preferences up to 20% more accurately than baselines.

Graph Convolutional Networks (GCNs) have achieved state-of-the-art performance on node classification. However, recent works have shown that GCNs are vulnerable to adversarial attacks, such as additions or deletions of adversarially-chosen edges in the graph, in order to mislead the node classification algorithms. How can we design robust GCNs that are resistant to such adversarial attacks? More challengingly, how can we do this in a way that is provably robust? We propose a robust node classification approach based on a low-pass `message passing' mechanism, that (a) reduces the effectiveness of adversarial attacks in experiments, and (b) provides theoretical guarantees against adversarial attacks. Our approach can be embedded into the existing GCN architectures to enhance their robustness. Empirical results show that our loss-pass method effectively improves the performance of multiple GCNs under miscellaneous perturbations and helps them to achieve superior performance on various graphs.

Graph data is ubiquitous in academia and industry, from social networks to bioinformatics. The pervasiveness of graphs today has raised the demand for algorithms that can answer various questions: Which products would a user like to purchase given her order list? Which users are buying fake followers to increase their public reputation? Myriads of new graph mining algorithms are proposed every year to answer such questions - each with a distinct problem formulation, computational time, and memory footprint. This lack of unity makes it difficult for a practitioner to compare different algorithms and pick the most suitable one for a specific application. These challenges - even more severe for non-experts - create a gap in which state-of-the-art techniques developed in academic settings fail to be optimally deployed in real-world applications. To bridge this gap, we propose AutoGM, an automated system for graph mining algorithm development. We first define a unified framework UnifiedGM that integrates various message-passing based graph algorithms, ranging from conventional algorithms like PageRank to graph neural networks. Then UnifiedGM defines a search space in which five parameters are required to determine a graph algorithm. Under this search space, AutoGM explicitly optimizes for the optimal parameter set of UnifiedGM using Bayesian Optimization. AutoGM defines a novel budget-aware objective function for the optimization to incorporate a practical issue - finding the best speed-accuracy trade-off under a computation budget - into the graph algorithm generation problem. Experiments on real-world benchmark datasets demonstrate that AutoGM generates novel graph mining algorithms with the best speed/accuracy trade-off compared to existing models with heuristic parameters.

Can one build a knowledge graph (KG) for all products in the world? Knowledge graphs have firmly established themselves as valuable sources of information for search and question answering, and it is natural to wonder if a KG can contain information about products offered at online retail sites. There have been several successful examples of generic KGs, but organizing information about products poses many additional challenges, including sparsity and noise of structured data for products, complexity of the domain with millions of product types and thousands of attributes, heterogeneity across large number of categories, as well as large and constantly growing number of products.

Taxonomies have found wide applications in various domains, especially online for item categorization, browsing, and search. Despite the prevalent use of online catalog taxonomies, most of them in practice are maintained by humans, which is labor-intensive and difficult to scale. While taxonomy construction from scratch is considerably studied in the literature, how to effectively enrich existing incomplete taxonomies remains an open yet important research question. Taxonomy enrichment not only requires the robustness to deal with emerging terms but also the consistency between existing taxonomy structure and new term attachment. In this paper, we present a self-supervised end-to-end framework, Octet, for Online Catalog Taxonomy EnrichmenT. Octet leverages heterogeneous information unique to online catalog taxonomies such as user queries, items, and their relations to the taxonomy nodes while requiring no other supervision than the existing taxonomies. We propose to distantly train a sequence labeling model for term extraction and employ graph neural networks (GNNs) to capture the taxonomy structure as well as the query-item-taxonomy interactions for term attachment. Extensive experiments in different online domains demonstrate the superiority of Octet over state-of-the-art methods via both automatic and human evaluations. Notably, Octet enriches an online catalog taxonomy in production to 2 times larger in the open-world evaluation.

Given multiple input signals, how can we infer node importance in a knowledge graph (KG)? Node importance estimation is a crucial and challenging task that can benefit a lot of applications including recommendation, search, and query disambiguation. A key challenge towards this goal is how to effectively use input from different sources. On the one hand, a KG is a rich source of information, with multiple types of nodes and edges. On the other hand, there are external input signals, such as the number of votes or pageviews, which can directly tell us about the importance of entities in a KG. While several methods have been developed to tackle this problem, their use of these external signals has been limited as they are not designed to consider multiple signals simultaneously. In this paper, we develop an end-to-end model MultiImport, which infers latent node importance from multiple, potentially overlapping, input signals. MultiImport is a latent variable model that captures the relation between node importance and input signals, and effectively learns from multiple signals with potential conflicts. Also, MultiImport provides an effective estimator based on attentive graph neural networks. We ran experiments on real-world KGs to show that MultiImport handles several challenges involved with inferring node importance from multiple input signals, and consistently outperforms existing methods, achieving up to 23.7% higher [email protected] than the state-of-the-art method.

Entity matching seeks to identify data records over one or multiple data sources that refer to the same real-world entity. Virtually every entity matching task on large datasets requires blocking, a step that reduces the number of record pairs to be matched. However, most of the traditional blocking methods are learning-free and key-based, and their successes are largely built on laborious human effort in cleaning data and designing blocking keys.

Do higher-order network structures aid graph semi-supervised learning? Given a graph and a few labeled vertices, labeling the remaining vertices is a high-impact problem with applications in several tasks, such as recommender systems, fraud detection and protein identification. However, traditional methods rely on edges for spreading labels, which is limited as all edges are not equal. Vertices with stronger connections participate in higher-order structures in graphs, which calls for methods that can leverage these structures in the semi-supervised learning tasks.

Knowledge graph (e.g. Freebase, YAGO) is a multi-relational graph representing rich factual information among entities of various types. Entity alignment is the key step towards knowledge graph integration from multiple sources. It aims to identify entities across different knowledge graphs that refer to the same real world entity. However, current entity alignment systems overlook the sparsity of different knowledge graphs and can not align multi-type entities by one single model. In this paper, we present a Collective Graph neural network for Multi-type entity Alignment, called CG-MuAlign. Different from previous work, CG-MuAlign jointly aligns multiple types of entities, collectively leverages the neighborhood information and generalizes to unlabeled entity types. Specifically, we propose novel collective aggregation function tailored for this task, that (1) relieves the incompleteness of knowledge graphs via both cross-graph and self attentions, (2) scales up efficiently with mini-batch training paradigm and effective neighborhood sampling strategy. We conduct experiments on real world knowledge graphs with millions of entities and observe the superior performance beyond existing methods. In addition, the running time of our approach is much less than the current state-of-the-art deep learning methods.

Given a stream of edge additions and deletions, how can we estimate the count of triangles in it? If we can store only a subset of the edges, how can we obtain unbiased estimates with small variances? Counting triangles (i.e., cliques of size three) in a graph is a classical problem with applications in a wide range of research areas, including social network analysis, data mining, and databases. Recently, streaming algorithms for triangle counting have been extensively studied since they can naturally be used for large dynamic graphs. However, existing algorithms cannot handle edge deletions or suffer from low accuracy. Can we handle edge deletions while achieving high accuracy? We propose T hink D, which accurately estimates the counts of global triangles (i.e., all triangles) and local triangles associated with each node in a fully dynamic graph stream with additions and deletions of edges. Compared to its best competitors, T hink D is (a) Accurate: up to 4.3 × more accurate within the same memory budget, (b) Fast: up to 2.2 × faster for the same accuracy requirements, and (c) Theoretically sound: always maintaining estimates with zero bias (i.e., the difference between the true triangle count and the expected value of its estimate) and small variance. As an application, we use T hink D to detect suddenly emerging dense subgraphs, and we show its advantages over state-of-the-art methods.

How can we early warn against an impending student drop out or an adverse health condition in near real-time? More challengingly, how do we learn to early warn from data containing confounding interventions-e.g., tutoring or medicines-while remaining interpretable to the human decision maker? We consider the problem of learning to interpretably early warn from labeled data tainted by interventions. We first identify three principles that an early warning system should follow. We then propose SmokeAlarm which provably obeys these principles and produces early warning scores in an online manner. Notably, learned model is "bi-inspectable", i.e., it can be visualized both in the presence and in the absence of interventions. Experiments demonstrate the efficacy of SmokeAlarm over prior approaches.

Time series forecasting is a key ingredient in the automation and optimization of business processes: in retail, deciding which products to order and where to store them depends on the forecasts of future demand in different regions; in cloud computing, the estimated future usage of services and infrastructure components guides capacity planning; and workforce scheduling in warehouses and factories requires forecasts of the future workload. Recent years have witnessed a paradigm shift in forecasting techniques and applications, from computer-assisted model- and assumption-based to data-driven and fully-automated. This shift can be attributed to the availability of large, rich, and diverse time series data sources and result in a set of challenges that need to be addressed such as the following. How can we build statistical models to efficiently and effectively learn to forecast from large and diverse data sources? How can we leverage the statistical power of "similar'' time series to improve forecasts in the case of limited observations? What are the implications for building forecasting systems that can handle large data volumes?

How can we estimate the importance of nodes in a knowledge graph (KG)? A KG is a multi-relational graph that has proven valuable for many tasks including question answering and semantic search. In this paper, we present GENI, a method for tackling the problem of estimating node importance in KGs, which enables several downstream applications such as item recommendation and resource allocation. While a number of approaches have been developed to address this problem for general graphs, they do not fully utilize information available in KGs, or lack flexibility needed to model complex relationship between entities and their importance. To address these limitations, we explore supervised machine learning algorithms. In particular, building upon recent advancement of graph neural networks (GNNs), we develop GENI, a GNN-based method designed to deal with distinctive challenges involved with predicting node importance in KGs. Our method performs an aggregation of importance scores instead of aggregating node embeddings via predicate-aware attention mechanism and flexible centrality adjustment. In our evaluation of GENI and existing methods on predicting node importance in real-world KGs with different characteristics, GENI achieves 5-17% higher [email protected] than the state of the art.

Given a dynamic graph stream, how can we detect the sudden appearance of anomalous patterns, such as link spam, follower boosting, or denial of service attacks? Additionally, can we categorize the types of anomalies that occur in practice, and theoretically analyze the anomalous signs arising from each type? In this work, we propose AnomRank, an online algorithm for anomaly detection in dynamic graphs. AnomRank uses a two-pronged approach defining two novel metrics for anomalousness. Each metric tracks the derivatives of its own version of a 'node score' (or node importance) function. This allows us to detect sudden changes in the importance of any node. We show theoretically and experimentally that the two-pronged approach successfully detects two common types of anomalies: sudden weight changes along an edge, and sudden structural changes to the graph. AnomRank is (a) Fast and Accurate: up to 49.5x faster or 35% more accurate than state-of-the-art methods, (b) Scalable: linear in the number of edges in the input graph, processing millions of edges within 2 seconds on a stock laptop/desktop, and (c) Theoretically Sound: providing theoretical guarantees of the two-pronged approach.

Given multivariate time series data, how do we detect changes in the behavior of the time series: for example, the onset of illnesses or complications in patients? Can we do this without making strong assumptions about the data? We propose BnB (Branch and Border), an online, nonparametric change detection method that detects multiple changes in multivariate data. Unlike existing methods, BnB approaches change detection by separating points before and after the change using an ensemble of random partitions. BnB is (a) scalable: it scales linearly in the number of time ticks and dimensions, and is online, thus using bounded memory and bounded time per iteration; (b) effective: providing theoretical guarantees on the false positive rate, and achieving 70% or more increased F-measure over baselines in experiments averaged over 11 datasets; (c) general: it is nonparametric, and works on mixed data, including numerical, categorical, and ordinal data.

Consider a stream of time-stamped events, such as taxi rides, where we record the start and end locations of each ride. How do we learn a matrix factorization model which takes into account seasonal patterns (such as: rides toward office areas occur more frequently in the morning), and use it to forecast taxi rides tomorrow? Also, how can we model drift (such as population growth), and detect sudden changes (or anomalies)? Existing matrix factorization algorithms do not take seasonal patterns into account. We propose SMF (Seasonal Matrix Factorization), a matrix factorization model for seasonal data, and a streaming algorithm for fitting it. SMF is (a) accurate in forecasting: outperforming baselines by 13% to 60% in RMSE; (b) online: requiring fixed memory even as more data is received over time, and scaling linearly; (c) effective: providing interpretable results. In addition, we propose SMF-A, an algorithm which detects anomalies in a computationally feasible way, without forecasting every observation in the matrix.

Time series forecasting is a key ingredient in the automation and optimization of business processes: in retail, deciding which products to order and where to store them depends on the forecasts of future demand in different regions; in cloud computing, the estimated future usage of services and infrastructure components guides capacity planning; and workforce scheduling in warehouses and factories requires forecasts of the future workload. Recent years have witnessed a paradigm shift in forecasting techniques and applications, from computer-assisted model- and assumption-based to data-driven and fully-automated. This shift can be attributed to the availability of large, rich, and diverse time series corpora and result in a set of challenges that need to be addressed such as the following. How can we build statistical models to efficiently and effectively learn to forecast from large and diverse data sources? How can we leverage the statistical power of "similar'' time series to improve forecasts in the case of limited observations? What are the implications for building forecasting systems that can handle large data volumes? The objective of this tutorial is to provide a concise and intuitive overview of the most important methods and tools available for solving large-scale forecasting problems. We review the state of the art in three related fields: (1) classical modeling of time series, (2) scalable tensor methods, and (3) deep learning for forecasting. Further, we share lessons learned from building scalable forecasting systems. While our focus is on providing an intuitive overview of the methods and practical issues which we will illustrate via case studies, we also present some technical details underlying these powerful tools.

No abstract available.

Generalized canonical correlation analysis (GCCA) integrates information from data samples that are acquired at multiple feature spaces (or `views') to produce low-dimensional representations-which is an extension of classical two-view CCA. Since the 1960s, (G)CCA has attracted much attention in statistics, machine learning, and data mining because of its importance in data analytics. Despite these efforts, the existing GCCA algorithms have serious complexity issues. The memory and computational complexities of the existing algorithms usually grow as a quadratic and cubic function of the problem dimension (the number of samples / features), respectively-e.g., handling views with ≈1,000 features using such algorithms already occupies ≈106memory and the periteration complexity is ≈109flops-which makes it hard to push these methods much further. To circumvent such difficulties, we first propose a GCCA algorithm whose memory and computational costs scale linearly in the problem dimension and the number of nonzero data elements, respectively. Consequently, the proposed algorithm can easily handle very large sparse views whose sample and feature dimensions both exceed 100,000. Our second contribution lies in proposing two distributed algorithms for GCCA, which compute the canonical components of different views in parallel and thus can further reduce the runtime significantly if multiple computing agents are available. We provide detailed convergence analyses of the proposed algorithms and show that all the largescale GCCA algorithms converge to a Karush-Kuhn-Tucker (KKT) point at least sublinearly. Judiciously designed synthetic and realdata experiments are employed to showcase the effectiveness of the proposed algorithms.

How can we detect fraud in a big graph with rich properties, as online fraudsters invest more resources, including purchasing large pools of fake user accounts and dedicated IPs, to hide their fraudulent attacks? To achieve robustness, existing approaches detected dense sub-graphs as suspicious patterns in an unsupervised way, such as average degree maximization. However, such approaches suffer from the bias of including more nodes than necessary, resulting in lower accuracy and increased need for manual verification. Therefore, we propose HoloScope, which introduces a novel metric “contrast suspiciousness” integrating information from graph topology and spikes to more accurately detect fraudulent users and objects. Contrast suspiciousness dynamically emphasizes the contrasting patterns between fraudsters and normal users, making HoloScope capable of distinguishing the synchronized and strange behaviors of fraudsters by means of topology, bursts and drops, and rating scores. In addition, we provide theoretical bounds for how much this method increases the time cost needed for fraudsters to conduct adversarial attacks. Moreover, HoloScope has a concise framework and sub-quadratic time complexity, making the algorithm reproducible and scalable. In extensive experiments, HoloScope achieved significant accuracy improvements on real data with injected labels and true labels, when compared with state-of-the-art fraud detection methods.

Rakshit Trivedi, Bunyamin Sisman, Xin Luna Dong, Christos Faloutsos, Jun Ma, Hongyuan Zha. Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2018.

Given electrical sensors placed on the power grid, how can we automatically determine when electrical components (e.g. power lines) fail? Or, given traffic sensors which measure the speed of vehicles passing over them, how can we determine when traffic accidents occur? Both these problems involve detecting change points in a set of sensors on the nodes or edges of a graph. To this end, we propose ChangeDAR (Change Detection And Resolution), which detects changes in an online manner, and reports when and where the change occurred in the graph.

More and more data need to be processed or analyzed within mobile devices for efficiency or privacy reasons, but performing machine learning tasks with large data within the devices is challenging because of their limited memory resources. For this reason, disk-based machine learning methods have been actively researched, which utilize storage resources without holding all the data in memory. This paper proposes D-MC2, a novel disk-based matrix completion method that (1) supports incremental data update (i.e., data insertion and deletion) and (2) spills both data and model to disk when necessary; these functionalities are not supported by existing methods. First, D-MC2 builds a two-layered index to efficiently support incremental data update; there exists a trade-off relationship between model learning and data update costs, and our two-layered index simultaneously optimizes the two costs. Second, we develop a window-based stochastic gradient descent (SGD) scheduler to efficiently support the dual spilling; a huge amount of disk I/O is incurred when the size of model is larger than that of memory, and our new scheduler substantially reduces it. Our evaluation results show that D-MC2 is significantly more scalable and faster than other disk-based competitors under the limited memory environment. In terms of the co-optimization, D-MC2 outperforms the baselines that only optimize one of the two costs up to 48x. Furthermore, the window-based scheduler improves the training speed 12.4x faster compared to a naive scheduler.

We address the challenge of reconstructing historical counts from aggregated, possibly overlapping historical reports. For example, given the monthly and weekly sums, how can we find the daily counts of people infected with flu? We propose an approach, called ARES (Automatic REStoration), that performs automatic data reconstruction in two phases: (1) first, it estimates the sequence of historical counts utilizing domain knowledge, such as smoothness and periodicity of historical events; (2) then, it uses the estimated sequence to learn notable patterns in the target sequence to refine the reconstructed time series. In order to derive such patterns, ARES uses an annihilating filter technique. The idea is to learn a linear shift-invariant operator whose response to the desired sequence is (approximately) zero-yielding a set of null-space equations that the desired signal should satisfy, without the need for the accompanying data. The reconstruction accuracy can be further improved by applying the second phase iteratively. We evaluate ARES on the real epidemiological data from the Tycho project and demonstrate that ARES recovers historical data from aggregated reports with high accuracy. In particular, it considerably outperforms top competitors, including least squares approximation and the more advanced H-FUSE method (42% and 34% improvement based on average RMSE, respectively).

Given a stream of edges from a time-evolving (un) weighted (un) directed graph, we consider the problem of detecting anomalous edges in near real-time using sublinear memory. We propose SedanSpot, a principled randomized algorithm, which exploits two tell-tale signs of anomalous edges: they tend to (i) occur as bursts of activity and (ii) connect parts of the graph which are sparsely connected. SedanSpot has the following desirable properties: (a) Burst Resistance: It provably downsamples edges from bursty periods of network traffic, (b) Holistic scoring: It takes into account the whole (sampled) graph while scoring the anomalousness of an edge, giving diminishing importance to far-away neighbors, (c) Efficiency: It supports fast updates and scoring and hence can be efficiently maintained over stream; further, it can detect anomalous edges in sublinear space and constant time per edge. Through experiments on real-world data, we demonstrate that SedanSpot is 3x faster and 270% more accurate (in terms of AUC) than the state-of-the-art.

Can we forecast future connections in a social network? Can we predict who will start using a given hashtag in Twitter, leveraging contextual information such as who follows or retweets whom to improve our predictions? In this paper we present an abridged report of TensorCast, an award winning method for forecasting time-evolving networks, that uses coupled tensors to incorporate multiple information sources. TensorCast is scalable (linearithmic on the number of connections), effective (more precise than competing methods) and general (applicable to any data source representable by a tensor). We also showcase our method when applied to forecast two large scale heterogeneous real world temporal networks, namely Twitter and DBLP.

How do we spot interesting events from e-mail or transportation logs? How can we detect port scan or denial of service attacks from IP-IP communication data? In general, given a sequence of weighted, directed or bipartite graphs, each summarizing a snapshot of activity in a time window, how can we spot anomalous graphs containing the sudden appearance or disappearance of large dense subgraphs (e.g., near bicliques) in near real-time using sublinear memory? To this end, we propose a randomized sketching-based approach called SpotLight, which guarantees that an anomalous graph is mapped 'far' away from 'normal' instances in the sketch space with high probability for appropriate choice of parameters. Extensive experiments on real-world datasets show that SpotLight (a) improves accuracy by at least 8.4% compared to prior approaches, (b) is fast and can process millions of edges within a few minutes, (c) scales linearly with the number of edges and sketching dimensions and (d) leads to interesting discoveries in practice.

Given a large, dynamic graph, how can we trace the activities of groups of vertices over time? Given a dynamic biological graph modeling a given disease progression, which genes interact closely at the early stage of the disease, and their interactions are being disrupted in the latter stage of the disease? Which genes interact sparsely at the early stage of the disease, and their interactions increase as the disease progresses? Knowing the answers to these questions is important as they give insights to the underlying molecular mechanism to disease progression, and potential treatments that target these mechanisms can be developed. There are three main contributions to this paper. First, we designed a novel algorithm, SDREGION, that identifies subgraphs that decrease or increase in density monotonically over time, referred to as d-regions or i-regions, respectively. We introduced the objective function, -density, for identifying d-(i-)regions. Second, SDREGION is a generic algorithm, applicable across several real datasets. In this manuscript, we showed its effectiveness, and made observations in the modeling of the progression of lung cancer. In particular, we observed that SDREGION identified d-(i-)regions that capture mechanisms that align with literature. Importantly, findings that were identified but were not retrospectively validated by literature may provide novel mechanisms in tumor progression that will guide future biological experiments. Third, SDREGION is scalable with a time complexity of O(mlogn + nlogn) where m is the number of edges, and n is the number of vertices in a given dynamic graph.

How can we efficiently forecast the power consumption of a location for the next few days? More challengingly, how can we forecast the power consumption if the temperature increases by 10° C, the number of appliances in the grid increase by 20%, and voltage levels increase by 5%? Such ‘what-if scenarios' are crucial for future planning, to ensure that the grid remains reliable even under extreme conditions. Our contributions are as follows: 1) Domain knowledge infusion: we propose a novel Temporal BIG model that extends the physics-based BIG model, allowing it to capture changes over time, trends, and seasonality, and temperature effects. 2) Forecasting: our StreamCast algorithm forecasts multiple steps ahead and outperforms baselines in accuracy. Our algorithm is online, requiring constant update time per new data point and bounded memory. 3) What-if scenarios and anomaly detection: our approach can handle scenarios in which the voltage levels, temperature, or number of appliances change. It also spots anomalies in real data, and provides confidence intervals for its forecasts, to assist in planning for various scenarios. Experimental results show that StreamCast has 27% lower forecasting error than baselines on real data, scales linearly, and runs in 4 minutes on a time sequence of 40 million points.

Rating platforms enable large-scale collection of user opinion about items(e.g., products or other users). However, untrustworthy users give fraudulent ratings for excessive monetary gains. In this paper, we present REV2, a system to identify such fraudulent users. We propose three interdependent intrinsic quality metrics---fairness of a user, reliability of a rating and goodness of a product. The fairness and reliability quantify the trustworthiness of a user and rating, respectively, and goodness quantifies the quality of a product. Intuitively, a user is fair if it provides reliable scores that are close to the goodness of products. We propose six axioms to establish the interdependency between the scores, and then, formulate a mutually recursive definition that satisfies these axioms. We extend the formulation to address cold start problem and incorporate behavior properties. We develop the REV2 algorithm to calculate these intrinsic quality scores for all users, ratings, and products. We show that this algorithm is guaranteed to converge and has linear time complexity. By conducting extensive experiments on five rating datasets, we show that REV2 outperforms nine existing algorithms in detecting fair and unfair users. We reported the 150 most unfair users in the Flipkart network to their review fraud investigators, and 127 users were identified as being fraudulent(84.6% accuracy). The REV2 algorithm is being deployed at Flipkart.

Recovering a time sequence of events from multiple aggregated and possibly overlapping reports is a major challenge in historical data fusion. The goal is to reconstruct a higher resolution event sequence from a mixture of lower resolution samples as accurately as possible. For example, we may aim to disaggregate overlapping monthly counts of people infected with measles into weekly counts. In this paper, we propose a novel data disaggregation method, called H ome R un , that exploits an alternative representation of the sequence and finds the spectrum of the target sequence. More specifically, we formulate the problem as so-called basis pursuit using the Discrete Cosine Transform (DCT) as a sparsifying dictionary and impose non-negativity and smoothness constraints. H ome R un utilizes the energy compaction feature of the DCT by finding the sparsest spectral representation of the target sequence that contains the largest (most important) coefficients. We leverage the Alternating Direction Method of Multipliers to solve the resulting optimization problem with scalable and memory efficient steps. Experiments using real epidemiological data show that our method considerably outperforms the state-of-the-art techniques, especially when the DCT of the sequence has a high degree of energy compaction.

Time series forecasting is a key ingredient in the automation and optimization of business processes: in retail, deciding which products to order and where to store them depends on the forecasts of future demand in different regions; in cloud computing, the estimated future usage of services and infrastructure components guides capacity planning; and workforce scheduling in warehouses, call centers, factories requires forecasts of the future workload. Recent years have witnessed a paradigm shift in forecasting techniques and applications, from computer-assisted model- and assumption-based to data-driven and fully-automated. This shift can be attributed to the availability of large, rich, and diverse time series data sources, posing unprecedented challenges to traditional time series forecasting methods. As such, how can we build statistical models to efficiently and effectively learn to forecast from large and diverse data sources? How can we leverage the statistical power of "similar" time series to improve forecasts in the case of limited observations? What are the implications for building forecasting systems that can handle large data volumes? The objective of this tutorial is to provide a concise and intuitive overview of the most important methods and tools available for solving large-scale forecasting problems. We review the state of the art in three related fields: (1) classical modeling of time series, (2) scalable tensor methods, and (3) deep learning for forecasting. Further, we share lessons learned from building scalable forecasting systems. While our focus is on providing an intuitive overview of the methods and practical issues, we also present technical details underlying these powerful tools.

Given a large-scale and high-order tensor, how can we detect dense subtensors in it? Can we spot them in near-linear time but with quality guarantees? Extensive previous work has shown that dense subtensors, as well as dense subgraphs, indicate anomalous or fraudulent behavior (e.g., lockstep behavior in social networks). However, available algorithms for detecting dense subtensors are not satisfactory in terms of speed, accuracy, and flexibility. In this work, we propose two algorithms, called M-Z oom and M-B iz , for fast and accurate dense-subtensor detection with various density measures. M-Z oom gives a lower bound on the density of detected subtensors, while M-B iz guarantees the local optimality of detected subtensors. M-Z oom and M-B iz can be combined, giving the following advantages: (1) Scalable: scale near-linearly with all aspects of tensors and are up to 114× faster than state-of-the-art methods with similar accuracy, (2) Provably accurate : provide a guarantee on the lowest density and local optimality of the subtensors they find, (3) Flexible: support multi-subtensor detection and size bounds as well as diverse density measures, and (4) Effective: successfully detected edit wars and bot activities in Wikipedia, and spotted network attacks from a TCP dump with near-perfect accuracy (AUC = 0.98).

What is the growth dynamics of social networks, like Facebook or WeChat? Does it truly exhibit exponential early-growth, as predicted by the celebrated models, like the Bass model? How about the dynamics of links, for which there are few published models? For the first time, we examine the growth of WeChat which is the largest online social network in China, together with several other real social networks. We observe Power-Law growth dynamics for both nodes and links, a fact that breaks the textbook models featuring Sigmoid curves. We propose NETTIDE, along with differential equations for the growth of nodes and links. Our model fits the growth dynamics of real social networks well; it encompasses many traditional growth dynamics as special cases, while remaining parsimonious in parameters. The NETTIDE for link growth is the first one of its kind, accurately fitting real data, and capturing densification phenomenon. We further formulate two stochastic generators, which interpret the growth of nodes and links through survival analysis and micro-level interactions within a social network, respectively. The proposed generators reproduce realistic growth dynamics of social networks. When applied on the WeChat data, our NETTIDE forecasted > 730 days ahead with 3 percent error.

As online fraudsters invest more resources, including purchasing large pools of fake user accounts and dedicated IPs, fraudulent attacks become less obvious and their detection becomes increasingly challenging. Existing approaches such as average degree maximization suffer from the bias of including more nodes than necessary, resulting in lower accuracy and increased need for manual verification. Hence, we propose HoloScope, which introduces a novel metric "contrast suspiciousness" integrating information from graph topology and spikes to more accurately detect fraudulent users and objects. Contrast suspiciousness dynamically emphasizes the contrast patterns between fraudsters and normal users, making HoloScope capable of distinguishing the synchronized and anomalous behaviors of fraudsters on topology, bursts and drops, and rating scores. In addition, we provide theoretical bounds for how much this increases the time cost needed for fraudsters to conduct adversarial attacks. Moreover, HoloScope has a concise framework and sub-quadratic time complexity, making the algorithm reproducible and scalable. Extensive experiments showed that HoloScope achieved significant accuracy improvements on synthetic and real data, compared with state-of-the-art fraud detection methods.

'Alice' is submitting one web search per five minutes, for three hours in a row-is it normal? How to detect abnormal searchbehaviors, among Alice and other users? Is there any distinct pattern in Alice's (or other users') search behavior? We studied what is probably the largest, publicly available, query log, containing more than 30 million queries from 0.6 million users. In this paper, we present a novel, user-and group-level framework, M3A: Model, MetaModel and Anomaly detection. For each user, we discover and explain a surprising, bi-modal pattern of the inter-arrival time (IAT) of landed queries (queries with user click-through). Specifically, the model Camel-Logis proposed to describe such an IAT distribution; we then notice the correlations among its parameters at the group level. Thus, we further propose the metamodel Meta-Click, to capture and explain the two-dimensional, heavy-tail distribution of the parameters. Combining Camel-Log and Meta-Click, the proposed M3A has the following strongpoints: (1) the accurate modeling of marginal IAT distribution, (2) quantitative interpretations, (3) anomaly detection, and (4) generality-being able to explain multiple datasets.

Large graph datasets have caused renewed interest for graph partitioning. However, existing well-studied graph partitioners often assume that vertices of the graph are always active during the computation, which may lead to time-varying skewness for traversal-style graph workloads, like Breadth First Search, since they only explore part of the graph in each superstep. Additionally, existing solutions do not consider what vertices each partition will have, as a result, high-degree vertices may be concentrated into a few partitions, causing imbalance. Towards this, we introduce the idea of skew-resistant graph partitioning, where the objective is to create an initial partitioning that will "hold well" over time without suffering from skewness. Skew-resistant graph partitioning tries to mitigate skewness by taking the characteristics of both the target workload and the graph structure into consideration.

Given an heterogeneous social network, can we forecast its future? Can we predict who will start using a given hashtag on twitter? Can we leverage side information, such as who retweets or follows whom, to improve our membership forecasts? We present TensorCast, a novel method that forecasts time-evolving networks more accurately than current state of the art methods by incorporating multiple data sources in coupled tensors. TensorCast is (a) scalable, being linearithmic on the number of connections; (b) effective, achieving over 20% improved precision on top-1000 forecasts of community members; (c) general, being applicable to data sources with different structure. We run our method on multiple real-world networks, including DBLP and a Twitter temporal network with over 310 million non-zeros, where we predict the evolution of the activity of the use of political hashtags.

How can we quickly explain large-scale directed and weighted graphs? We present Spectral Lens (SL) to analyze a variety of real-world networks with both negative and positive edge weights, like DBLP relationships, email communications and Bitcoin trust votings. SL offers value on three levels: (a) Diagnostics: Spectral Lens combines spectral properties from Singular Value Decomposition to create an SL-Dictionary (SLD) to enhance understanding of any directed, weighted graph. (b) Tools: the SL-Algorithm (SLA) automatically extracts the top groups of nodes with similar connectivity, finds groups of shared connectivity and detects suspicious behaviour in a graph. (c) Discoveries: Experiments on several real-world networks illustrate the effectiveness of SLA. Observations from synthetic and real-world networks reveal relations between spectral and graph properties. We show that SLA is highly scalable and linear on the size of a graph. Analyzing a graph with over 2 million edges takes less than 5 minutes. Overall, SL provides an easy-to-use tool for practitioners to explain a weighted and directed graph quickly, to understand its connectivity and identify regular and anomalous behaviors.

Most past work on social network link fraud detection tries to separate genuine users from fraudsters, implicitly assuming that there is only one type of fraudulent behavior. But is this assumption true? And, in either case, what are the characteristics of such fraudulent behaviors? In this work, we set up honeypots ("dummy" social network accounts), and buy fake followers (after careful IRB approval). We report the signs of such behaviors including oddities in local network connectivity, account attributes, and similarities and differences across fraud providers. Most valuably, we discover and characterize several types of fraud behaviors. We discuss how to leverage our insights in practice by engineering strongly performing entropy-based features and demonstrating high classification accuracy. Our contributions are (a) observations: we analyze our honeypot fraudster ecosystem and give surprising insights into the multifaceted behaviors of these fraudster types, and (b) features: we propose novel features that give strong (>0.95 precision/recall) discriminative power on ground-truth Twitter data.

Given a retinal image, can we automatically determine whether it is of high quality (suitable for medical diagnosis)? Can we also explain our decision, pinpointing the region or regions that led to our decision? Images from human retinas are vital for the diagnosis of multiple health issues, like hypertension, diabetes, and Alzheimer's; low quality images may force the patient to come back again for a second scanning, wasting time and possibly delaying treatment. However, existing retinal image quality assessment methods are either black boxes without explanations of the results or depend heavily on feature engineering or on complex and error-prone anatomical structures' segmentation. Therefore, we propose EyeQual, that solves exactly this problem. EyeQual is novel, fast for inference, accurate and explainable, pinpointing low-quality regions on the image. We evaluated EyeQual on two real datasets where it achieved 100% accuracy taking just 36 milliseconds for each image.

How can we design a product or movie that will attract, for example, the interest of Pennsylvania adolescents or liberal newspaper critics? What should be the genre of that movie and who should be in the cast? In this work, we seek to identify how we can design new movies with features tailored to a specific user population. We formulate the movie design as an optimization problem over the inference of user-feature scores and selection of the features that maximize the number of attracted users. Our approach, PNP, is based on a heterogeneous, tripartite graph of users, movies, and features (e.g. actors, directors, genres), where users rate movies and features contribute to movies. We learn the preferences by leveraging user similarities defined through different types of relations, and show that our method outperforms state-of-the-art approaches, including matrix factorization and other heterogeneous graph-based analysis. We evaluate PNP on publicly available real-world data and show that it is highly scalable and effectively provides movie designs oriented towards different groups of users, including men, women, and adolescents.

Consider a stream of retweet events - how can we spot fraudulent lock-step behavior in such multi-aspect data (i.e., tensors) evolving over time? Can we detect it in real time, with an accuracy guarantee? Past studies have shown that dense subtensors tend to indicate anomalous or even fraudulent behavior in many tensor data, including social media, Wikipedia, and TCP dumps. Thus, several algorithms have been proposed for detecting dense subtensors rapidly and accurately. However, existing algorithms assume that tensors are static, while many real-world tensors, including those mentioned above, evolve over time.

How do people make friends dynamically in social networks? What are the temporal patterns for an individual increasing its social connectivity? What are the basic mechanisms governing the formation of these temporal patterns? No matter cyber or physical social systems, their structure and dynamics are mainly driven by the connectivity dynamics of each individual. However, due to the lack of empirical data, little is known about the empirical dynamic patterns of social connectivity at microscopic level, let alone the regularities or models governing these microscopic dynamics.

Is it possible to spot review frauds and spamming on social media and online stores? In this paper we analyze the joint distribution of the inter-arrival times and volume of events such as comments and online reviews and show that it is possible to accurately rank and detect suspicious users such as spammers, bots and fraudsters. We propose VolTime, a generative model that fits well the inter-arrival time distribution (IAT) of real users. Thus, VOLTIME automatically spots and ranks suspicious users. Experiments on several real datasets, ranging from Reddit comments and phone calls to Flipkart product reviews, show that VolTime is able to accurately fit the activity volume and IAT of real data. Additionally, we show that VolTime ranks suspicious users with a precision higher than 90% for a sensitivity of 70%.

Given a friendship network, how certain are we that Smith is a progressive (vs. conservative)? How can we propagate these certainties through the network? While Belief propagation marked the beginning of principled label-propagation to classify nodes in a graph, its numerous variants proposed in the literature fail to take into account uncertainty during the propagation process. As we show, this limitation leads to counter-intuitive results for even simple graphs. Motivated by these observations, we formalize axioms that any node classification algorithm should obey and propose NetConf which satisfies these axioms and handles arbitrary network effects (homophily/heterophily) at scale. Our contributions are: (1) Axioms: We state axioms that any node classification algorithm should satisfy; (2) Theory: NetConf is grounded in a Bayesian-theoretic framework to model uncertainties, has a closed-form solution and comes with precise convergence guarantees; (3) Practice: Our method is easy to implement and scales linearly with the number of edges in the graph. On experiments using real world data, we always match or outperform BP while taking less processing time.

How close can we zoom in to observe brain activity? Our understanding is limited by the resolution of imaging modalities that exhibit good spatial but poor temporal resolution, or vice-versa. In this paper, we propose BrainZoom, an efficient imaging algorithm that cross-leverages multi-modal brain signals. BrainZoom (a) constructs high resolution brain images from multi-modal signals, (b) is scalable, and (c) is flexible in that it can easily incorporate various priors on the brain activities, such as sparsity, low rank, or smoothness. We carefully formulate the problem to tackle nonlinearity in the measurements (via variable splitting) and auto-scale between different modal signals, and judiciously design an inexact alternating optimization-based algorithmic framework to handle the problem with provable convergence guarantees. Our experiments using a popular realistic brain signal simulator to generate fMRI and MEG demonstrate that high spatio-temporal resolution brain imaging is possible from these two modalities. The experiments also suggest that smoothness seems to be the best prior, among several we tried.

In this paper, we address the challenge of recovering a time sequence of counts from aggregated historical data. For example, given a mixture of the monthly and weekly sums, how can we find the daily counts of people infected with flu? In general, what is the best way to recover historical counts from aggregated, possibly overlapping historical reports, in the presence of missing values? Equally importantly, how much should we trust this reconstruction? We propose H-Fuse, a novel method that solves above problems by allowing injection of domain knowledge in a principled way, and turning the task into a well-defined optimization problem. H-Fuse has the following desirable properties: (a) Effectiveness, recovering historical data from aggregated reports with high accuracy; (b) Self-awareness, providing an assessment of when the recovery is not reliable; (c) Scalability, computationally linear on the size of the input data. Experiments on the real data (epidemiology counts from the Tycho project [13]) demonstrates that H-FUSE reconstructs the original data 30 – 81% better than the least squares method.

Multi-aspect data appear frequently in many web-related applications. For example, product reviews are quadruplets of (user, product, keyword, timestamp). How can we analyze such web-scale multi-aspect data? Can we analyze them on an off-the-shelf workstation with limited amount of memory?

How can we detect fraudulent lockstep behavior in large-scale multi-aspect data (i.e., tensors)? Can we detect it when data are too large to fit in memory or even on a disk? Past studies have shown that dense blocks in real-world tensors (e.g., social media, Wikipedia, TCP dumps, etc.) signal anomalous or fraudulent behavior such as retweet boosting, bot activities, and network attacks. Thus, various approaches, including tensor decomposition and search, have been used for rapid and accurate dense-block detection in tensors. However, all such methods have low accuracy, or assume that tensors are small enough to fit in main memory, which is not true in many real-world applications such as social media and web.

Given a collection of seasonal time-series, how can we find regular (cyclic) patterns and outliers (i.e. rare events)? These two types of patterns are hidden and mixed in the time-varying activities. How can we robustly separate regular patterns and outliers, without requiring any prior information?

Given a heterogeneous network, with nodes of different types - e.g., products, users and sellers from an online recommendation site like Amazon - and labels for a few nodes ('honest', 'suspicious', etc), can we find a closed formula for Belief Propagation (BP), exact or approximate? Can we say whether it will converge? BP, traditionally an inference algorithm for graphical models, exploits so-called "network effects" to perform graph classification tasks when labels for a subset of nodes are provided; and it has been successful in numerous settings like fraudulent entity detection in online retailers and classification in social networks. However, it does not have a closed-form nor does it provide convergence guarantees in general. We propose ZooBP, a method to perform fast BP on undirected heterogeneous graphs with provable convergence guarantees. ZooBP has the following advantages: (1) Generality : It works on heterogeneous graphs with multiple types of nodes and edges; (2) Closed-form solution: ZooBP gives a closed-form solution as well as convergence guarantees; (3) Scalability: ZooBP is linear on the graph size and is up to 600× faster than BP, running on graphs with 3.3 million edges in a few seconds. (4) Effectiveness: Applied on real data (a F lipkart e-commerce network with users, products and sellers), ZooBP identifies fraudulent users with a near-perfect precision of 92.3 % over the top 300 results.

Tensors and tensor decompositions are very powerful and versatile tools that can model a wide variety of heterogeneous, multiaspect data. As a result, tensor decompositions, which extract useful latent information out of multiaspect data tensors, have witnessed increasing popularity and adoption by the data mining community. In this survey, we present some of the most widely used tensor decompositions, providing the key insights behind them, and summarizing them from a practitioner’s point of view. We then provide an overview of a very broad spectrum of applications where tensors have been instrumental in achieving state-of-the-art performance, ranging from social network analysis to brain data analysis, and from web mining to healthcare. Subsequently, we present recent algorithmic advances in scaling tensor decompositions up to today’s big data, outlining the existing systems and summarizing the key ideas behind them. Finally, we conclude with a list of challenges and open problems that outline exciting future research directions.

Social media has become a popular and important tool for human communication. However, due to this popularity, spam and the distribution of malicious content by computer-controlled users, known as bots, has become a widespread problem. At the same time, when users use social media, they generate valuable data that can be used to understand the patterns of human communication. In this article, we focus on the following important question: Can we identify and use patterns of human communication to decide whether a human or a bot controls a user? The first contribution of this article is showing that the distribution of inter-arrival times (IATs) between postings is characterized by following four patterns: (i) heavy-tails, (ii) periodic-spikes, (iii) correlation between consecutive values, and (iv) bimodallity. As our second contribution, we propose a mathematical model named Act-M (Activity Model). We show that Act-M can accurately fit the distribution of IATs from social media users. Finally, we use Act-M to develop a method that detects if users are bots based only on the timing of their postings. We validate Act-M using data from over 55 million postings from four social media services: Reddit, Twitter, Stack-Overflow, and Hacker-News. Our experiments show that Act-M provides a more accurate fit to the data than existing models for human dynamics. Additionally, when detecting bots, Act-M provided a precision higher than 93% and 77% with a sensitivity of 70% for the Twitter and Reddit datasets, respectively.

Given a bipartite graph of users and the products that they review, or followers and followees, how can we detect fake reviews or follows? Existing fraud detection methods (spectral, etc.) try to identify dense subgraphs of nodes that are sparsely connected to the remaining graph. Fraudsters can evade these methods using camouflage , by adding reviews or follows with honest targets so that they look “normal.” Even worse, some fraudsters use hijacked accounts from honest users, and then the camouflage is indeed organic. Our focus is to spot fraudsters in the presence of camouflage or hijacked accounts. We propose FRAUDAR, an algorithm that (a) is camouflage resistant, (b) provides upper bounds on the effectiveness of fraudsters, and (c) is effective in real-world data. Experimental results under various attacks show that FRAUDAR outperforms the top competitor in accuracy of detecting both camouflaged and non-camouflaged fraud. Additionally, in real-world experiments with a Twitter follower--followee graph of 1.47 billion edges, FRAUDAR successfully detected a subgraph of more than 4, 000 detected accounts, of which a majority had tweets showing that they used follower-buying services.

How do social groups, such as Facebook groups and Wechat groups, dynamically evolve over time? How do people join the social groups, uniformly or with burst? What is the pattern of people quitting from groups? Is there a simple universal model to depict the come-and-go patterns of various groups? In this article, we examine temporal evolution patterns of more than 100 thousands social groups with more than 10 million users. We surprisingly find that the evolution patterns of real social groups goes far beyond the classic dynamic models like SI and SIR. For example, we observe both diffusion and non-diffusion mechanism in the group joining process, and power-law decay in group quitting process, rather than exponential decay as expected in SIR model. Therefore, we propose a new modelcomeNgo, a concise yet flexible dynamic model for group evolution. Our model has the following advantages: (a) Unification power: it generalizes earlier theoretical models and different joining and quitting mechanisms we find from observation. (b) Succinctness and interpretability: it contains only six parameters with clear physical meanings. (c) Accuracy: it can capture various kinds of group evolution patterns preciously, and the goodness of fit increases by 58% over baseline. (d) Usefulness: it can be used in multiple application scenarios, such as forecasting and pattern discovery. Furthermore, our model can provide insights about different evolution patterns of social groups, and we also find that group structure and its evolution has notable relations with temporal patterns of group evolution.

The recent explosion in the adoption of search engines and new media such as blogs and Twitter have facilitated the faster propagation of news and rumors. How quickly does a piece of news spread over these media? How does its popularity diminish over time? Does the rising and falling pattern follow a simple universal law? In this article, we propose SpikeM, a concise yet flexible analytical model of the rise and fall patterns of information diffusion. Our model has the following advantages. First, unification power: it explains earlier empirical observations and generalizes theoretical models including the SI and SIR models. We provide the threshold of the take-off versus die-out conditions for SpikeM and discuss the generality of our model by applying it to an arbitrary graph topology. Second, practicality: it matches the observed behavior of diverse sets of real data. Third, parsimony: it requires only a handful of parameters. Fourth, usefulness: it makes it possible to perform analytic tasks such as forecasting, spotting anomalies, and interpretation by reverse engineering the system parameters of interest (quality of news, number of interested bloggers, etc.). We also introduce an efficient and effective algorithm for the real-time monitoring of information diffusion, namely SpikeStream, which identifies multiple diffusion patterns in a large collection of online event streams. Extensive experiments on real datasets demonstrate that SpikeM accurately and succinctly describes all patterns of the rise and fall spikes in social networks.

We address the problem of estimating core numbers of nodes by reading edges of a large graph stored in external memory. The core number of a node is the highest k-core in which the node participates. Core numbers are useful in many graph mining tasks, especially ones that involve finding communities of nodes, influential spreaders and dense subgraphs. Large graphs often do not fit on the memory of a single machine. Existing external memory solutions do not give bounds on the required space. In practice, existing solutions also do not scale with the size of the graph. We propose NimbleCore, an iterative external-memory algorithm, which estimates core numbers of nodes using O(n log dmax) space, where n is the number of nodes and dmaxis the maximum node-degree in the graph. We also show that NimbleCore requires O(n) space for graphs with power-law degree distributions. Experiments on forty-eight large graphs from various domains demonstrate that NimbleCore gives space savings up to 60X, while accurately estimating core numbers with average relative error less than 2.3%.

Given "who-trusts/distrusts-whom" information, how can we propagate the trust and distrust? With the appearance of fraudsters in social network sites, the importance of trust prediction has increased. Most such methods use only explicit and implicit trust information (e.g., if Smith likes several of Johnson's reviews, then Smith implicitly trusts Johnson), but they do not consider distrust.

In social voting Web sites, how do the user actions - up-votes, down-votes and comments - evolve over time? Are there relationships between votes and comments? What is normal and what is suspicious? These are the questions we focus on. We analyzed over 20,000 submissions corresponding to more than 100 million user interactions from three social voting Web sites: Reddit, Imgur and Digg. Our first contribution is two discoveries: (i) the number of comments grows as a power-law on the number of votes and (ii) the time between a submission creation and a user's reaction obeys a log-logistic distribution. Based on these patterns, we propose VnC (Vote-and-Comment), a parsimonious but accurate and scalable model that models the coevolution of user activities. In our experiments on real data, VnC outperformed state-of-the-art baselines on accuracy. Additionally, we illustrate VnC usefulness for forecasting and outlier detection.

Generalized canonical correlation analysis (GCCA) aims at extracting common structure from multiple 'views', i.e., high-dimensional matrices representing the same objects in different feature domains – an extension of classical two-view CCA. Existing (G)CCA algorithms have serious scalability issues, since they involve square root factorization of the correlation matrices of the views. The memory and computational complexity associated with this step grow as a quadratic and cubic function of the problem dimension (the number of samples / features), respectively. To circumvent such difficulties, we propose a GCCA algorithm whose memory and computational costs scale linearly in the problem dimension and the number of nonzero data elements, respectively. Consequently, the proposed algorithm can easily handle very large sparse views whose sample and feature dimensions both exceed 100,000 – while the current approaches can only handle thousands of features / samples. Our second contribution is a distributed algorithm for GCCA, which computes the canonical components of different views in parallel and thus can further reduce the runtime significantly (by ≥ 30% in experiments) if multiple cores are available. Judiciously designed synthetic and real-data experiments using a multilingual dataset are employed to showcase the effectiveness of the proposed algorithms.

Weighted signed networks (WSNs) are networks in which edges are labeled with positive and negative weights. WSNs can capture like/dislike, trust/distrust, and other social relationships between people. In this paper, we consider the problem of predicting the weights of edges in such networks. We propose two novel measures of node behavior: the goodness of a node intuitively captures how much this node is liked/trusted by other nodes, while the fairness of a node captures how fair the node is in rating other nodes' likeability or trust level. We provide axioms that these two notions need to satisfy and show that past work does not meet these requirements for WSNs. We provide a mutually recursive definition of these two concepts and prove that they converge to a unique solution in linear time. We use the two measures to predict the edge weight in WSNs. Furthermore, we show that when compared against several individual algorithms from both the signed and unsigned social network literature, our fairness and goodness metrics almost always have the best predictive power. We then use these as features in different multiple regression models and show that we can predict edge weights on 2 Bitcoin WSNs, an Epinions WSN, 2 WSNs derived from Wikipedia, and a WSN derived from Twitter with more accurate results than past work. Moreover, fairness and goodness metrics form the most significant feature for prediction in most (but not all) cases.

How do the k-core structures of real-world graphs look like? What are the common patterns and the anomalies? How can we use them for algorithm design and applications? A k-core is the maximal subgraph where all vertices have degree at least k. This concept has been applied to such diverse areas as hierarchical structure analysis, graph visualization, and graph clustering. Here, we explore pervasive patterns that are related to k-cores and emerging in graphs from several diverse domains. Our discoveries are as follows: (1) Mirror Pattern: coreness of vertices (i.e., maximum k such that each vertex belongs to the k-core) is strongly correlated to their degree. (2) Core-Triangle Pattern: degeneracy of a graph (i.e., maximum k such that the k-core exists in the graph) obeys a 3-to-1 power law with respect to the count of triangles. (3) Structured Core Pattern: degeneracy-cores are not cliques but have non-trivial structures such as core-periphery and communities. Our algorithmic contributions show the usefulness of these patterns. (1) Core-A, which measures the deviation from Mirror Pattern, successfully finds anomalies in real-world graphs complementing densest-subgraph based anomaly detection methods. (2) Core-D, a single-pass streaming algorithm based on Core-Triangle Pattern, accurately estimates the degeneracy of billion-scale graphs up to 7× faster than a recent multipass algorithm.(3) Core-S, inspired by Structured Core Pattern, identifies influential spreaders up to 17× faster than top competitors with comparable accuracy.

Given a bipartite graph of users and the products that they review, or followers and followees, how can we detect fake reviews or follows? Existing fraud detection methods (spectral, etc.) try to identify dense subgraphs of nodes that are sparsely connected to the remaining graph. Fraudsters can evade these methods using camouflage, by adding reviews or follows with honest targets so that they look "normal". Even worse, some fraudsters use hijacked accounts from honest users, and then the camouflage is indeed organic. Our focus is to spot fraudsters in the presence of camouflage or hijacked accounts. We propose FRAUDAR, an algorithm that (a) is camouflage-resistant, (b) provides upper bounds on the effectiveness of fraudsters, and (c) is effective in real-world data. Experimental results under various attacks show that FRAUDAR outperforms the top competitor in accuracy of detecting both camouflaged and non-camouflaged fraud. Additionally, in real-world experiments with a Twitter follower-followee graph of 1.47 billion edges, FRAUDAR successfully detected a subgraph of more than 4000 detected accounts, of which a majority had tweets showing that they used follower-buying services.

Representing and summarizing human behaviors with rich contexts facilitates behavioral sciences and user-oriented services. Traditional behavioral modeling represents a behavior as a tuple in which each element is one contextual factor of one type, and the tensor-based summaries look for high-order dense blocks by clustering the values (including timestamps) in each dimension. However, the human behaviors are multicontextual and dynamic: (1) each behavior takes place within multiple contexts in a few dimensions, which requires the representation to enable non-value and set-values for each dimension; (2) many behavior collections, such as tweets or papers, evolve over time. In this paper, we represent the behavioral data as a two-level matrix (temporal-behaviors by dimensional-values) and propose a novel representation for behavioral summary called Tartan that includes a set of dimensions, the values in each dimension, a list of consecutive time slices and the behaviors in each slice. We further develop a propagation method CatchTartan to catch the dynamic multicontextual patterns from the temporal multidimensional data in a principled and scalable way: it determines the meaningfulness of updating every element in the Tartan by minimizing the encoding cost in a compression manner. CatchTartan outperforms the baselines on both the accuracy and speed. We apply CatchTartan to four Twitter datasets up to 10 million tweets and the DBLP data, providing comprehensive summaries for the events, human life and scientific development.

What is the growth pattern of social networks, like Facebook and WeChat? Does it truly exhibit exponential early growth, as predicted by textbook models like the Bass model, SI, or the Branching Process? How about the count of links, over time, for which there are few published models?

How do social groups, such as Facebook groups and Wechat groups, dynamically evolve over time? How do people join the social groups, uniformly or with burst? What is the pattern of people quitting from groups? Is there a simple universal model to depict the come-and-go patterns of various groups?

Review fraud is a pervasive problem in online commerce, in which fraudulent sellers write or purchase fake reviews to manipulate perception of their products and services. Fake reviews are often detected based on several signs, including 1) they occur in short bursts of time; 2) fraudulent user accounts have skewed rating distributions. However, these may both be true in any given dataset. Hence, in this paper, we propose an approach for detecting fraudulent reviews which combines these 2 approaches in a principled manner, allowing successful detection even when one of these signs is not present. To combine these 2 approaches, we formulate our Bayesian Inference for Rating Data (BIRD) model, a flexible Bayesian model of user rating behavior. Based on our model we formulate a likelihood-based suspiciousness metric, Normalized Expected Surprise Total (NEST). We propose a linear-time algorithm for performing Bayesian inference using our model and computing the metric. Experiments on real data show that BIRDNEST successfully spots review fraud in large, real-world graphs: the 50 most suspicious users of the Flipkart platform flagged by our algorithm were investigated and all identified as fraudulent by domain experts at Flipkart.

How do we model and find outliers in Twitter data? Given the number of retweets of each person on a social network, what is their expected number of comments? Real-life data are often very skewed, exhibiting power-law-like behavior. For such skewed multidimensional discrete data, the existing models are not general enough to capture various realistic scenarios, and need to be discretized as they often model continuous quantities. We propose FusionRP, short for Fusion Restaurant Process, a simple and intuitive model for skewed multi-dimensional discrete distributions, such as number of retweets vs. comments in Twitter-like data. Our model is discrete by design, has provably asymptotic log-logistic sum of marginals, is general enough to capture varied relationships, and most importantly, fits real data very well. We give an effective and scalable maximum-likelihood based fitting approach that is linear in the number of unique observed values and the input dimension. We test FusionRP on a twitter-like social network with 2.2M users, a phone call network with 1.9M call records, game data with 45M users and Facebook data with 2.5M posts. Our results show that FusionRP significantly outperforms several alternative methods and can detect outliers, such as bot-like behaviors in the Facebook data.

How can we tell if Alice is a talkative person or a silent person? In this paper, we focus on the node classification problem on networked data such as social networks and the web. There are two open challenges with this problem: (1) we want to handle various kinds of label correlations in real-world networks such as homophily (i.e., love of the same) and heterophily (i.e., love of the different), and (2) we want to exploit the confidence of the inference results to enhance the accuracy. There is no algorithm that solves these two challenges at the same time. We tackle with these two challenges by proposing CAMLP, a novel node classification algorithm. Our contributions are three-fold: (a) Novel algorithm; our algorithm is confidence-aware and is applicable to both homophily and heterophily networks, (b) Theory; we give theoretical analyses of our algorithm, and (c) Practice; we perform extensive experiments on 5 different network datasets including homophily and heterophily networks. Our experiments show that the proposed algorithm improves the precision of major competitors not only on heterophily networks, but also on homophily networks.

Which song will Smith listen to next? Which restaurant will Alice go to tomorrow? Which product will John click next? These applications have in common the prediction of user trajectories that are in a constant state of flux over a hidden network (e.g. website links, geographic location). Moreover, what users are doing now may be unrelated to what they will be doing in an hour from now. Mindful of these challenges we propose TribeFlow, a method designed to cope with the complex challenges of learning personalized predictive models of non-stationary, transient, and time-heterogeneous user trajectories. TribeFlow is a general method that can perform next product recommendation, next song recommendation, next location prediction, and general arbitrary-length user trajectory prediction without domain-specific knowledge. TribeFlow is more accurate and up to 413x faster than top competitors.

Given a large collection of time-evolving activities, such as Google search queries, which consist of d keywords/activities for m locations of duration n, how can we analyze temporal patterns and relationships among all these activities and find location-specific trends? How do we go about capturing non-linear evolutions of local activities and forecasting future patterns? For example, assume that we have the online search volume for multiple keywords, e.g., "Nokia/Nexus/Kindle" or "CNN/BBC" for 236 countries/territories, from 2004 to 2015. Our goal is to analyze a large collection of multi-evolving activities, and specifically, to answer the following questions: (a) Is there any sign of interaction/competition between two different keywords? If so, who competes with whom? (b) In which country is the competition strong? (c) Are there any seasonal/annual activities? (d) How can we automatically detect important world-wide (or local) events? We present COMPCUBE, a unifying non-linear model, which provides a compact and powerful representation of co-evolving activities; and also a novel fitting algorithm, COMPCUBE-FIT, which is parameter-free and scalable. Our method captures the following important patterns: (B)asic trends, i.e., non-linear dynamics of co-evolving activities, signs of (C)ompetition and latent interaction, e.g., Nokia vs. Nexus, (S)easonality, e.g., a Christmas spike for iPod in the U.S. and Europe, and (D)eltas, e.g., unrepeated local events such as the U.S. election in 2008. Thanks to its concise but effective summarization, COMPCUBE can also forecast long-range future activities. Extensive experiments on real datasets demonstrate that COMPCUBE consistently outperforms the best state-of- the-art methods in terms of both accuracy and execution speed.

Large graphs are prevalent in many applications and enable a variety of information dissemination processes, e.g., meme, virus, and influence propagation. How can we optimize the underlying graph structure to affect the outcome of such dissemination processes in a desired way (e.g., stop a virus propagation, facilitate the propagation of a piece of good idea, etc)? Existing research suggests that the leading eigenvalue of the underlying graph is the key metric in determining the so-called epidemic threshold for a variety of dissemination models. In this paper, we study the problem of how to optimally place a set of edges (e.g., edge deletion and edge addition) to optimize the leading eigenvalue of the underlying graph, so that we can guide the dissemination process in a desired way. We propose effective, scalable algorithms for edge deletion and edge addition, respectively. In addition, we reveal the intrinsic relationship between edge deletion and node deletion problems. Experimental results validate the effectiveness and efficiency of the proposed algorithms.

Mining knowledge from a multimedia database has received increasing attentions recently since huge repositories are made available by the development of the Internet. In this article, we exploit the relations among different modalities in a multimedia database and present a framework for general multimodal data mining problem where image annotation and image retrieval are considered as the special cases. Specifically, the multimodal data mining problem can be formulated as a structured prediction problem where we learn the mapping from an input to the structured and interdependent output variables. In addition, in order to reduce the demanding computation, we propose a new max margin structure learning approach called Enhanced Max Margin Learning (EMML) framework, which is much more efficient with a much faster convergence rate than the existing max margin learning methods, as verified through empirical evaluations. Furthermore, we apply EMML framework to develop an effective and efficient solution to the multimodal data mining problem that is highly scalable in the sense that the query response time is independent of the database scale. The EMML framework allows an efficient multimodal data mining query in a very large scale multimedia database, and excels many existing multimodal data mining methods in the literature that do not scale up at all. The performance comparison with a state-of-the-art multimodal data mining method is reported for the real-world image databases.

Given a directed graph of millions of nodes, how can we automatically spot anomalous, suspicious nodes judging only from their connectivity patterns? Suspicious graph patterns show up in many applications, from Twitter users who buy fake followers, manipulating the social network, to botnet members performing distributed denial of service attacks, disturbing the network traffic graph. We propose a fast and effective method, C atch S ync , which exploits two of the tell-tale signs left in graphs by fraudsters: (a) synchronized behavior: suspicious nodes have extremely similar behavior patterns because they are often required to perform some task together (such as follow the same user); and (b) rare behavior: their connectivity patterns are very different from the majority. We introduce novel measures to quantify both concepts (“synchronicity” and “normality”) and we propose a parameter-free algorithm that works on the resulting synchronicity-normality plots. Thanks to careful design, C atch S ync has the following desirable properties: (a) it is scalable to large datasets, being linear in the graph size; (b) it is parameter free ; and (c) it is side-information-oblivious : it can operate using only the topology, without needing labeled data, nor timing information, and the like., while still capable of using side information if available. We applied C atch S ync on three large, real datasets, 1-billion-edge Twitter social graph, 3-billion-edge, and 12-billion-edge Tencent Weibo social graphs, and several synthetic ones; C atch S ync consistently outperforms existing competitors, both in detection accuracy by 36% on Twitter and 20% on Tencent Weibo, as well as in speed.

How much has a network changed since yesterday? How different is the wiring of Bob’s brain (a left-handed male) and Alice’s brain (a right-handed female), and how is it different? Graph similarity with given node correspondence, i.e., the detection of changes in the connectivity of graphs, arises in numerous settings. In this work, we formally state the axioms and desired properties of the graph similarity functions, and evaluate when state-of-the-art methods fail to detect crucial connectivity changes in graphs. We propose D elta C on , a principled, intuitive, and scalable algorithm that assesses the similarity between two graphs on the same nodes (e.g., employees of a company, customers of a mobile carrier). In conjunction, we propose D elta C on -A ttr , a related approach that enables attribution of change or dissimilarity to responsible nodes and edges. Experiments on various synthetic and real graphs showcase the advantages of our method over existing similarity measures. Finally, we employ D elta C on and D elta C on -A ttr on real applications: (a) we classify people to groups of high and low creativity based on their brain connectivity graphs, (b) do temporal anomaly detection in the who-emails-whom Enron graph and find the top culprits for the changes in the temporal corporate email graph, and (c) recover pairs of test-retest large brain scans ( ∼17M edges, up to 90M edges) for 21 subjects.

Given a large graph, like a computer communication network, which k nodes should we immunize (or monitor, or remove), to make it as robust as possible against a computer virus attack? This problem, referred to as the node immunization problem, is the core building block in many high-impact applications, ranging from public health, cybersecurity to viral marketing. A central component in node immunization is to find the best k bridges of a given graph. In this setting, we typically want to determine the relative importance of a node (or a set of nodes) within the graph, for example, how valuable (as a bridge) a person or a group of persons is in a social network. First of all, we propose a novel `bridging' score Dλ, inspired by immunology, and we show that its results agree with intuition for several realistic settings. Since the straightforward way to compute Dλ is computationally intractable, we then focus on the computational issues and propose a surprisingly efficient way (O(nk2+ m)) to estimate it. Experimental results on real graphs show that (1) the proposed `bridging' score gives mining results consistent with intuition; and (2) the proposed fast solution is up to seven orders of magnitude faster than straightforward alternatives.

Many commercial products and academic research activities are embracing behavior analysis as a technique for improving detection of attacks of many sorts-from retweet boosting, hashtag hijacking to link advertising. Traditional approaches focus on detecting dense blocks in the adjacency matrix of graph data, and recently, the tensors of multimodal data. No method gives a principled way to score the suspiciousness of dense blocks with different numbers of modes and rank them to draw human attention accordingly. In this paper, we first give a list of axioms that any metric of suspiciousness should satisfy; we propose an intuitive, principled metric that satisfies the axioms, and is fast to compute; moreover, we propose CrossSpot, an algorithm to spot dense blocks that are worth inspecting, typically indicating fraud or some other noteworthy deviation from the usual, and sort them in the order of importance (“suspiciousness”). Finally, we apply CrossSpot to the real data, where it improves the F1 score over previous techniques by 68 percent and finds suspicious behavioral patterns in social datasets spanning 0.3 billion posts.

If we know most of Smith’s friends are from Boston, what can we say about the rest of Smith’s friends? In this paper, we focus on the node classification problem on networks, which is one of the most important topics in AI and Web communities. Our proposed algorithm which is referred to as OMNIProp has the following properties: (a) seamless and accurate; it works well on any label correlations (i.e., homophily, heterophily, and mixture of them) (b) fast; it is efficient and guaranteed to converge on arbitrary graphs (c) quasi-parameter free; it has just one well-interpretable parameter with heuristic default value of 1. We also prove the theoretical connections of our algorithm to the semi-supervised learning (SSL) algorithms and to random-walks. Experiments on four real, different network datasets demonstrate the benefits of the proposed algorithm, where OMNI-Prop outperforms the top competitors.

How do people interact with their Facebook wall? At a high level, this question captures the essence of our work. While most prior efforts focus on Twitter, the much fewer Facebook studies focus on the friendship graph or are limited by the amount of users or the duration of the study. In this work, we model Facebook user behavior: we analyze the wall activities of users focusing on identifying common patterns and surprising phenomena. We conduct an extensive study of roughly 7K users over three years during four month intervals each year. We propose PowerWall, a lesser known heavy-tailed distribution to fit our data. Our key results can be summarized in the following points. First, we find that many wall activities, including number of posts, number of likes, number of posts of type photo, etc., can be described by the PowerWall distribution. What is more surprising is that most of these distributions have similar slope, with a value close to 1! Second, we show how our patterns and metrics can help us spot surprising behaviors and anomalies. For example, we find a user posting every two days, exactly the same count of posts; another user posting at midnight, with no other activity before or after. Our work provides a solid step towards a systematic and quantitative wall-centric profiling of Facebook user activity.

A multi-faceted graph defines several facets on a set of nodes. Each facet is a set of edges that represent the relationships between the nodes in a specific context. Mining multi-faceted graphs have several applications, including finding fraudster rings that launch advertising traffic fraud attacks, tracking IP addresses of botnets over time, analyzing interactions on social networks and co-authorship of scientific papers. We propose NeSim, a distributed efficient clustering algorithm that does soft clustering on individual facets. We also propose optimizations to further improve the scalability, the efficiency and the clusters quality. We employ generalpurpose graph-clustering algorithms in a novel way to discover communities across facets. Due to the qualities of NeSim, we employ it as a backbone in the distributed MuFace algorithm, which discovers multi-faceted communities. We evaluate the proposed algorithms on several real and synthetic datasets, where NeSim is shown to be superior to MCL, JP and AP, the well-established clustering algorithms. We also report the success stories of MuFace in finding advertisement click rings.

How do anomalies, fraud, and spam effect our models of normal user behavior? How can we modify our models to catch fraudsters? In this tutorial we will answer these questions - connecting graph analysis tools for user behavior modeling to anomaly and fraud detection. In particular, we will focus on three data mining techniques: subgraph analysis, label propagation and latent factor models; and their application to static graphs, e.g. social networks, evolving graphs, e.g. "who-calls-whom" networks, and attributed graphs, e.g. the "who-reviews-what" graphs of Amazon and Yelp. For each of these techniques we will give an explanation of the algorithms and the intuition behind them. We will then give brief examples of recent research using the techniques to model, understand and predict normal behavior. With this intuition for how these methods are applied to graphs and user behavior, we will focus on state-of-the-art research showing how the outcomes of these methods are effected by fraud, and how they have been used to catch fraudsters.

Kejun Huang, Matt Gardner, Evangelos Papalexakis, Christos Faloutsos, Nikos Sidiropoulos, Tom Mitchell, Partha P. Talukdar, Xiao Fu. Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing. 2015.

Multilinear analysis is pervasive in a wide variety of fields, ranging from Signal Processing to Chemometrics, and from Machine Vision to Data Mining. Determining the quality of a given tensor decomposition is a task of utmost importance that spans all fields of application of tensors. This task by itself is hard in its nature, since even determining the rank of a tensor is an NP-hard problem. Fortunately, there exist heuristics in the literature that can be effectively used for this task; one of these heuristics is the so-called Core Consistency Diagnostic (CORCONDIA) which is very intuitive and simple. However simple, computation of this diagnostic proves to be a very daunting task even for data of medium scale, let alone big tensor data. With the increase of the size of the tensor data that need to be analyzed there grows the need for efficient and scalable algorithms to compute diagnostics such as CORCONDIA, in order to assess the modelling quality. In this work we derive a fast and exact algorithm for CORCONDIA which exploits data sparsity and scales very well as the tensor size increases.

How can we find useful patterns and anomalies in large scale real-world data with multiple attributes? For example, network intrusion logs, with (source-ip, target-ip, port-number, timestamp)? Tensors are suitable for modeling these multi-dimensional data, and widely used for the analysis of social networks, web data, network traffic, and in many other settings. However, current tensor decomposition methods do not scale for tensors with millions and billions of rows, columns and `fibers', that often appear in real datasets. In this paper, we propose HaTen2, a scalable distributed suite of tensor decomposition algorithms running on the MapReduce platform. By carefully reordering the operations, and exploiting the sparsity of real world tensors, HaTen2 dramatically reduces the intermediate data, and the number of jobs. As a result, using HaTen2, we analyze big real-world tensors that can not be handled by the current state of the art, and discover hidden concepts.

Which seems more suspicious: 5,000 tweets from 200 users on 5 IP addresses, or 10,000 tweets from 500 users on 500 IP addresses but all with the same trending topic and all in 10 minutes? The literature has many methods that try to find dense blocks in matrices, and, recently, tensors, but no method gives a principled way to score the suspiciouness of dense blocks with different numbers of modes and rank them to draw human attention accordingly. Dense blocks are worth inspecting, typically indicating fraud, emerging trends, or some other noteworthy deviation from the usual. Our main contribution is that we show how to unify these methods and how to give a principled answer to questions like the above. Specifically, (a) we give a list of axioms that any metric of suspicousness should satisfy, (b) we propose an intuitive, principled metric that satisfies the axioms, and is fast to compute, (c) we propose CROSSSPOT, an algorithm to spot dense regions, and sort them in importance ("suspiciousness") order. Finally, we apply CROSSSPOT to real data, where it improves the F1 score over previous techniques by 68% and finds retweet-boosting in a real social dataset spanning 0.3 billion posts.

Given a large bipartite graph that represents objects and their properties, how can we automatically extract semantic information that provides an overview of the data and -- at the same time -- enables us to drill down to specific parts for an in-depth analysis? In this work, we propose extracting a taxonomy that models the relation between the properties via an is a hierarchy. The extracted taxonomy arranges the properties from general to specific providing different levels of abstraction. Our proposed method has the following desirable properties: (a) it requires no user-defined parameters, by exploiting the principle of minimum description length, (b) it is effective, by utilizing the inheritance of objects when representing the hierarchy, and (c) it is scalable, being linear in the number of edges. We demonstrate the effectiveness and scalability of our method on a broad spectrum of real, publicly available graphs from drug-property graphs to social networks with up to 22 million vertices and 286 million edges.

How can we model users' preferences? How do anomalies, fraud, and spam effect our models of normal users? How can we modify our models to catch fraudsters? In this tutorial we will answer these questions - connecting graph analysis tools for user behavior modeling to anomaly and fraud detection. In particular, we will focus on the application of subgraph analysis, label propagation, and latent factor models to static, evolving, and attributed graphs. For each of these techniques we will give a brief explanation of the algorithms and the intuition behind them. We will then give examples of recent research using the techniques to model, understand and predict normal behavior. With this intuition for how these methods are applied to graphs and user behavior, we will focus on state-of-the-art research showing how the outcomes of these methods are effected by fraud, and how they have been used to catch fraudsters.

Can we identify patterns of temporal activities caused by human communications in social media? Is it possible to model these patterns and tell if a user is a human or a bot based only on the timing of their postings? Social media services allow users to make postings, generating large datasets of human activity time-stamps. In this paper we analyze time-stamp data from social media services and find that the distribution of postings inter-arrival times (IAT) is characterized by four patterns: (i) positive correlation between consecutive IATs, (ii) heavy tails, (iii) periodic spikes and (iv) bimodal distribution. Based on our findings, we propose Rest-Sleep-and-Comment (RSC), a generative model that is able to match all four discovered patterns. We demonstrate the utility of RSC by showing that it can accurately fit real time-stamp data from Reddit and Twitter. We also show that RSC can be used to spot outliers and detect users with non-human behavior, such as bots. We validate RSC using real data consisting of over 35 million postings from Twitter and Reddit. RSC consistently provides a better fit to real data and clearly outperform existing models for human dynamics. RSC was also able to detect bots with a precision higher than 94%.

How can we describe a large, dynamic graph over time? Is it random? If not, what are the most apparent deviations from randomness -- a dense block of actors that persists over time, or perhaps a star with many satellite nodes that appears with some fixed periodicity? In practice, these deviations indicate patterns -- for example, botnet attackers forming a bipartite core with their victims over the duration of an attack, family members bonding in a clique-like fashion over a difficult period of time, or research collaborations forming and fading away over the years. Which patterns exist in real-world dynamic graphs, and how can we find and rank them in terms of importance? These are exactly the problems we focus on in this work. Our main contributions are (a) formulation: we show how to formalize this problem as minimizing the encoding cost in a data compression paradigm, (b) algorithm: we propose TIMECRUNCH, an effective, scalable and parameter-free method for finding coherent, temporal patterns in dynamic graphs and (c) practicality: we apply our method to several large, diverse real-world datasets with up to 36 million edges and 6.3 million nodes. We show that TIMECRUNCH is able to compress these graphs by summarizing important temporal structures and finds patterns that agree with intuition.

How can we reverse-engineer the brain connectivity, given the input stimulus, and the corresponding brain-activity measurements, for several experiments? We show how to solve the problem in a principled way, modeling the brain as a linear dynamical system (LDS), and solving the resulting “system identification” problem after imposing sparsity and non-negativity constraints on the appropriate matrices. These are reasonable assumptions in some applications, including magnetoencephalography (MEG). There are three contributions: (a) Proof: We prove that this simple condition resolves the ambiguity of similarity transformation in the LDS identification problem; (b) Algorithm: we propose an effective algorithm which further induces sparse connectivity in a principled way; and (c) Validation: our experiments on semi-synthetic (C. elegans), as well as real MEG data, show that our method recovers the neural connectivity, and it leads to interpretable results.

Given a large collection of time series, such as web-click logs, electric medical records and motion capture sensors, how can we efficiently and effectively find typical patterns? How can we statistically summarize all the sequences, and achieve a meaningful segmentation? What are the major tools for forecasting and outlier detection? Time-series data analysis is becoming of increasingly high importance, thanks to the decreasing cost of hardware and the increasing on-line processing capability.

How does a new startup drive the popularity of competing websites into oblivion like Facebook famously did to MySpace? This question is of great interest to academics, technologists, and financial investors alike. In this work we exploit the singular way in which Facebook wiped out the popularity of MySpace, Hi5, Friendster, and Multiply to guide the design of a new popularity competition model. Our model provides new insights into what Nobel Laure- ate Herbert A. Simon called the "marketplace of attention," which we recast as the attention-activity marketplace. Our model design is further substantiated by user-level activity of 250,000 MySpace users obtained between 2004 and 2009. The resulting model not only accurately fits the observed Daily Active Users (DAU) of Facebook and its competitors but also predicts their fate four years into the future.

Given a large collection of co-evolving online activities, such as searches for the keywords "Xbox", "PlayStation" and "Wii", how can we find patterns and rules? Are these keywords related? If so, are they competing against each other? Can we forecast the volume of user activity for the coming month? We conjecture that online activities compete for user attention in the same way that species in an ecosystem compete for food. We present ECOWEB, (i.e., Ecosystem on the Web), which is an intuitive model designed as a non-linear dynamical system for mining large-scale co-evolving online activities. Our second contribution is a novel, parameter-free, and scalable fitting algorithm, ECOWEB-FIT, that estimates the parameters of ECOWEB. Extensive experiments on real data show that ECOWEB is effective, in that it can capture long-range dynamics and meaningful patterns such as seasonalities, and practical, in that it can provide accurate long-range forecasts. ECOWEB consistently outperforms existing methods in terms of both accuracy and execution speed.

How can we tell when accounts are fake or real in a social network? And how can we tell which accounts belong to liberal, conservative or centrist users? Often, we can answer such questions and label nodes in a network based on the labels of their neighbors and appropriate assumptions of homophily ("birds of a feather flock together") or heterophily ("opposites attract"). One of the most widely used methods for this kind of inference is Belief Propagation (BP) which iteratively propagates the information from a few nodes with explicit labels throughout a network until convergence. A well-known problem with BP, however, is that there are no known exact guarantees of convergence in graphs with loops. This paper introduces Linearized Belief Propagation (LinBP), a linearization of BP that allows a closed-form solution via intuitive matrix equations and, thus, comes with exact convergence guarantees. It handles homophily, heterophily, and more general cases that arise in multi-class settings. Plus, it allows a compact implementation in SQL. The paper also introduces Single-pass Belief Propagation (SBP), a localized (or "myopic") version of LinBP that propagates information across every edge at most once and for which the final class assignments depend only on the nearest labeled neighbors. In addition, SBP allows fast incremental updates in dynamic networks. Our runtime experiments show that LinBP and SBP are orders of magnitude faster than standard BP, while leading to almost identical node labels.

Given a large graph with several millions or billions of nodes and edges, such as a social network, how can we explore it efficiently and find out what is in the data? In this demo we present P erseus , a large-scale system that enables the comprehensive analysis of large graphs by supporting the coupled summarization of graph properties and structures, guiding attention to outliers, and allowing the user to interactively explore normal and anomalous node behaviors. Specifically, P erseus provides for the following operations: 1) It automatically extracts graph invariants ( e.g. , degree, PageRank, real eigenvectors) by performing scalable, offline batch processing on H adoop ; 2) It interactively visualizes univariate and bivariate distributions for those invariants; 3) It summarizes the properties of the nodes that the user selects; 4) It efficiently visualizes the induced subgraph of a selected node and its neighbors, by incrementally revealing its neighbors. In our demonstration, we invite the audience to interact with P erseus to explore a variety of multi-million-edge social networks including a Wikipedia vote network, a friendship/foeship network in Slashdot, and a trust network based on the consumer review website Epinions.com.

With the advancement of information systems, means of communications are becoming cheaper, faster, and more available. Today, millions of people carrying smartphones or tablets are able to communicate practically any time and anywhere they want. They can access their e-mails, comment on weblogs, watch and post videos and photos (as well as comment on them), and make phone calls or text messages almost ubiquitously. Given this scenario, in this article, we tackle a fundamental aspect of this new era of communication: How the time intervals between communication events behave for different technologies and means of communications. Are there universal patterns for the Inter-Event Time Distribution (IED)? How do inter-event times behave differently among particular technologies? To answer these questions, we analyzed eight different datasets from real and modern communication data and found four well-defined patterns seen in all the eight datasets. Moreover, we propose the use of the Self-Feeding Process (SFP) to generate inter-event times between communications. The SFP is an extremely parsimonious point process that requires at most two parameters and is able to generate inter-event times with all the universal properties we observed in the data. We also show three potential applications of the SFP: as a framework to generate a synthetic dataset containing realistic communication events of any one of the analyzed means of communications, as a technique to detect anomalies, and as a building block for more specific models that aim to encompass the particularities seen in each of the analyzed systems.

How can we efficiently decompose a tensor into sparse factors, when the data do not fit in memory? Tensor decompositions have gained a steadily increasing popularity in data-mining applications; however, the current state-of-art decomposition algorithms operate on main memory and do not scale to truly large datasets. In this work, we propose P ar C ube , a new and highly parallelizable method for speeding up tensor decompositions that is well suited to produce sparse approximations. Experiments with even moderately large data indicate over 90% sparser outputs and 14 times faster execution, with approximation error close to the current state of the art irrespective of computation and memory requirements. We provide theoretical guarantees for the algorithm’s correctness and we experimentally validate our claims through extensive experiments, including four different real world datasets (E nron , L bnl , F acebook and N ell ), demonstrating its effectiveness for data-mining practitioners. In particular, we are the first to analyze the very large N ell dataset using a sparse tensor decomposition, demonstrating that P ar C ube enables us to handle effectively and efficiently very large datasets. Finally, we make our highly scalable parallel implementation publicly available, enabling reproducibility of our work.

A parallel algorithm for low-rank tensor decomposition that is especially well-suited for big tensors is proposed. The new algorithm is based on parallel processing of a set of randomly compressed, reduced-size `replicas' of the big tensor. Each replica is independently decomposed, and the results are joined via a master linear equation per tensor mode. The approach enables massive parallelism with guaranteed identifiability properties: if the big tensor has low rank and the system parameters are appropriately chosen, then the rank-one factors of the big tensor will be exactly recovered from the analysis of the reduced-size replicas. The proposed algorithm is proven to yield memory / storage and complexity gains of order up to IJ/F for a big tensor of size I × J × K of rank F with F ≤I ≤J ≤K.

If Lisa visits Dr. Brown, and there is no record of the drug he prescribed her, can we find it? Data sources, much to analysts' dismay, are too often plagued with incompleteness, making business analytics over the data difficult. Data entries with incomplete values are ignored, making some analytic queries fail to accurately describe how an organization is performing. We introduce a principled way of performing value imputation on missing values, allowing a user to choose a correct value after viewing possible values and why they were inferred. We achieve this by turning our data into a graph network and performing link prediction on nodes of interest using the belief propagation algorithm.

How can we detect suspicious users in large online networks? Online popularity of a user or product (via follows, page-likes, etc.) can be monetized on the premise of higher ad click-through rates or increased sales. Web services and social networks which incentivize popularity thus suffer from a major problem of fake connections from link fraudsters looking to make a quick buck. Typical methods of catching this suspicious behavior use spectral techniques to spot large groups of often blatantly fraudulent (but sometimes honest) users. However, small-scale, stealthy attacks may go unnoticed due to the nature of low-rank Eigen analysis used in practice. In this work, we take an adversarial approach to find and prove claims about the weaknesses of modern, state-of-the-art spectral methods and propose fBox, an algorithm designed to catch small-scale, stealth attacks that slip below the radar. Our algorithm has the following desirable properties: (a) it has theoretical underpinnings, (b) it is shown to be highly effective on real data and (c) it is scalable (linear on the input size). We evaluate fBox on a large, public 41.7 million node, 1.5 billion edge who-follows-whom social graph from Twitter in 2010 and with high precision identify many suspicious accounts which have persisted without suspension even to this day.

Rating data is ubiquitous on websites such as Amazon, TripAdvisor, or Yelp. Since ratings are not static but given at various points in time, a temporal analysis of rating data provides deeper insights into the evolution of a product's quality. In this work, we tackle the following question: Given the time stamped rating data for a product or service, how can we detect the general rating behavior of users as well as time intervals where the ratings behave anomalous? We propose a Bayesian model that represents the rating data as sequence of categorical mixture models. In contrast to existing methods, our method does not require any aggregation of the input but it operates on the original time stamped data. To capture the dynamic effects of the ratings, the categorical mixtures are temporally constrained: Anomalies can occur in specific time intervals only and the general rating behavior should evolve smoothly over time. Our method automatically determines the intervals where anomalies occur, and it captures the temporal effects of the general behavior by using a state space model on the natural parameters of the categorical distributions. For learning our model, we propose an efficient algorithm combining principles from variational inference and dynamic programming. In our experimental study we show the effectiveness of our method and we present interesting discoveries on multiple real world datasets.

Given a directed graph of millions of nodes, how can we automatically spot anomalous, suspicious nodes, judging only from their connectivity patterns? Suspicious graph patterns show up in many applications, from Twitter users who buy fake followers, manipulating the social network, to botnet members performing distributed denial of service attacks, disturbing the network traffic graph. We propose a fast and effective method, CatchSync, which exploits two of the tell-tale signs left in graphs by fraudsters: (a) synchronized behavior: suspicious nodes have extremely similar behavior pattern, because they are often required to perform some task together (such as follow the same user); and (b) rare behavior: their connectivity patterns are very different from the majority. We introduce novel measures to quantify both concepts ("synchronicity" and "normality") and we propose a parameter-free algorithm that works on the resulting synchronicity-normality plots. Thanks to careful design, CatchSync has the following desirable properties: (a) it is scalable to large datasets, being linear on the graph size; (b) it is parameter free; and (c) it is side-information-oblivious: it can operate using only the topology, without needing labeled data, nor timing information, etc., while still capable of using side information, if available. We applied CatchSync on two large, real datasets 1-billion-edge Twitter social graph and 3-billion-edge Tencent Weibo social graph, and several synthetic ones; CatchSync consistently outperforms existing competitors, both in detection accuracy by 36% on Twitter and 20% on Tencent Weibo, as well as in speed.

Given a large collection of epidemiological data consisting of the count of d contagious diseases for l locations of duration n, how can we find patterns, rules and outliers? For example, the Project Tycho provides open access to the count infections for U.S. states from 1888 to 2013, for 56 contagious diseases (e.g., measles, influenza), which include missing values, possible recording errors, sudden spikes (or dives) of infections, etc. So how can we find a combined model, for all these diseases, locations, and time-ticks? In this paper, we present FUNNEL, a unifying analytical model for large scale epidemiological data, as well as a novel fitting algorithm, FUNNELFIT, which solves the above problem. Our method has the following properties: (a) Sense-making: it detects important patterns of epidemics, such as periodicities, the appearance of vaccines, external shock events, and more; (b) Parameter-free: our modeling framework frees the user from providing parameter values; (c) Scalable: FUNNELFIT is carefully designed to be linear on the input size; (d) General: our model is general and practical, which can be applied to various types of epidemics, including computer-virus propagation, as well as human diseases. Extensive experiments on real data demonstrate that FUNNELFIT does indeed discover important properties of epidemics: (P1) disease seasonality, e.g., influenza spikes in January, Lyme disease spikes in July and the absence of yearly periodicity for gonorrhea; (P2) disease reduction effect, e.g., the appearance of vaccines; (P3) local/state-level sensitivity, e.g., many measles cases in NY; (P4) external shock events, e.g., historical flu pandemics; (P5) detect incongruous values, i.e., data reporting errors.

Given a simple noun such as {\em apple}, and a question such as "is it edible?", what processes take place in the human brain? More specifically, given the stimulus, what are the interactions between (groups of) neurons (also known as functional connectivity) and how can we automatically infer those interactions, given measurements of the brain activity? Furthermore, how does this connectivity differ across different human subjects?

Given multiple data sets of relational data that share a number of dimensions, how can we efficiently decompose our data into the latent factors? Factorization of a single matrix or tensor has attracted much attention, as, e.g., in the Netflix challenge, with users rating movies. However, we often have additional, side, information, like, e.g., demographic data about the users, in the Netflix example above. Incorporating the additional information leads to the coupled factorization problem. So far, it has been solved for relatively small datasets. We provide a distributed, scalable method for decomposing matrices, tensors, and coupled data sets through stochastic gradient descent on a variety of objective functions. We offer the following contributions: (1) Versatility: Our algorithm can perform matrix, tensor, and coupled factorization, with flexible objective functions including the Frobenius norm, Frobenius norm with an ℒ1 induced sparsity, and non-negative factorization. (2) Scalability: FlexiFaCT scales to unprecedented sizes in both the data and model, with up to billions of parameters. FlexiFaCT runs on standard Hadoop. (3) Convergence proofs showing that Flexi-FaCT converges on the variety of objective functions, even with projections.

How can we succinctly describe a million-node graph with a few simple sentences? How can we measure the ‘importance’ of a set of discovered subgraphs in a large graph? These are exactly the problems we focus on. Our main ideas are to construct a ‘vocabulary’ of subgraph-types that often occur in real graphs (e.g., stars, cliques, chains), and from a set of subgraphs, find the most succinct description of a graph in terms of this vocabulary. We measure success in a well-founded way by means of the Minimum Description Length (MDL) principle: a subgraph is included in the summary if it decreases the total description length of the graph. Our contributions are three-fold: (a) formulation: we provide a principled encoding scheme to choose vocabulary subgraphs; (b) algorithm: we develop VOG, an efficient method to minimize the description cost, and (c) applicability: we report experimental results on multi-million-edge real graphs, including Flickr and the Notre Dame web graph.

How can we correlate the neural activity in the human brain as it responds to typed words, with properties of these terms (like ‘edible’, ‘fits in hand’)? In short, we want to find latent variables, that jointly explain both the brain activity, as well as the behavioral responses. This is one of many settings of the Coupled Matrix-Tensor Factorization (CMTF) problem. Can we accelerate any CMTF solver, so that it runs within a few minutes instead of tens of hours to a day, while maintaining good accuracy? We introduce Turbo-SMT, a meta-method capable of doing exactly that: it boosts the performance of any CMTF algorithm, by up to 200x, along with an up to 65 fold increase in sparsity, with comparable accuracy to the baseline. We apply Turbo-SMT to BrainQ, a dataset consisting of a (nouns, brain voxels, human subjects) tensor and a (nouns, properties) matrix, with coupling along the nouns dimension. Turbo-SMT is able to find meaningful latent variables, as well as to predict brain activity with competitive accuracy.

Given a large collection of co-evolving multiple time-series, which contains an unknown number of patterns of different durations, how can we efficiently and effectively find typical patterns and the points of variation? How can we statistically summarize all the sequences, and achieve a meaningful segmentation? In this paper we present AutoPlait, a fully automatic mining algorithm for co-evolving time sequences. Our method has the following properties: (a) effectiveness: it operates on large collections of time-series, and finds similar segment groups that agree with human intuition; (b) scalability: it is linear with the input size, and thus scales up very well; and (c) AutoPlait is parameter-free, and requires no user intervention, no prior training, and no parameter tuning. Extensive experiments on 67GB of real datasets demonstrate that AutoPlait does indeed detect meaningful patterns correctly, and it outperforms state-of-the-art competitors as regards accuracy and speed: AutoPlait achieves near-perfect, over 95% precision and recall, and it is up to 472 times faster than its competitors.

Given a large dataset of users' ratings of movies, what is the best model to accurately predict which movies a person will like? And how can we prevent spammers from tricking our algorithms into suggesting a bad movie? Is it possible to infer structure between movies simultaneously? In this paper we describe a unified Bayesian approach to Collaborative Filtering that accomplishes all of these goals. It models the discrete structure of ratings and is flexible to the often non-Gaussian shape of the distribution. Additionally, our method finds a co-clustering of the users and items, which improves the model's accuracy and makes the model robust to fraud. We offer three main contributions: (1) We provide a novel model and Gibbs sampling algorithm that accurately models the quirks of real world ratings, such as convex ratings distributions. (2) We provide proof of our model's robustness to spam and anomalous behavior. (3) We use several real world datasets to demonstrate the model's effectiveness in accurately predicting user's ratings, avoiding prediction skew in the face of injected spam, and finding interesting patterns in real world ratings data.

Given a large graph, like who-calls-whom, or who-likes-whom, what behavior is normal and what should be surprising, possibly due to fraudulent activity? How do graphs evolve over time? How does influence/news/viruses propagate, over time? We focus on three topics: (a) anomaly detection in large static graphs (b) patterns and anomalies in large time-evolving graphs and (c) cascades and immunization. For the first, we present a list of static and temporal laws, including advances patterns like 'eigenspokes'; we show how to use them to spot suspicious activities, in on-line buyer-and-seller settings, in FaceBook, in twitter-like networks. For the second, we show how to handle time-evolving graphs as tensors, how to handle large tensors in map-reduce environments, as well as some discoveries such settings. For the third, we show that for virus propagation, a single number is enough to characterize the connectivity of graph, and thus we show how to do efficient immunization for almost any type of virus (SIS - no immunity; SIR - lifetime immunity; etc) We conclude with some open research questions for graph mining.

User provided rating data about products and services is one key feature of websites such as Amazon, TripAdvisor, or Yelp. Since these ratings are rather static but might change over time, a temporal analysis of rating distributions provides deeper insights into the evolution of a products' quality. Given a time-series of rating distributions, in this work, we answer the following questions: (1) How to detect the base behavior of users regarding a product's evaluation over time? (2) How to detect points in time where the rating distribution differs from this base behavior, e.g., due to attacks or spontaneous changes in the product's quality? To achieve these goals, we model the base behavior of users regarding a product as a latent multivariate autoregressive process. This latent behavior is mixed with a sparse anomaly signal finally leading to the observed data. We propose an efficient algorithm solving our objective and we present interesting findings on various real world datasets.

Let us consider that someone is starting a research on a topic that is unfamiliar to them. Which seminal papers have influenced the topic the most? What is the genealogy of the seminal papers in this topic? These are the questions that they can raise, which we try to answer in this paper. First, we propose an algorithm that finds a set of seminal papers on a given topic. We also address the performance and scalability issues of this sophisticated algorithm. Next, we discuss the measures to decide how much a paper is influenced by another paper. Then, we propose an algorithm that constructs a genealogy of the seminal papers by using the influence measure and citation information. Finally, through extensive experiments with a large volume of a real-world academic literature data, we show the effectiveness and efficiency of our approach.

Given a graph with billions of nodes and edges, how can we find patterns and anomalies? Are there nodes that participate in too many or too few triangles? Are there close-knit near-cliques? These questions are expensive to answer unless we have the first several eigenvalues and eigenvectors of the graph adjacency matrix. However, eigensolvers suffer from subtle problems (e.g., convergence) for large sparse matrices, let alone for billion-scale ones. We address this problem with the proposed HEIGEN algorithm, which we carefully design to be accurate, efficient, and able to run on the highly scalable MAPREDUCE (HADOOP) environment. This enables HEIGEN to handle matrices more than 1;000 × larger than those which can be analyzed by existing algorithms. We implement HEIGEN and run it on the M45 cluster, one of the top 50 supercomputers in the world. We report important discoveries about nearcliques and triangles on several real-world graphs, including a snapshot of the Twitter social network (56 Gb, 2 billion edges) and the “YahooWeb” data set, one of the largest publicly available graphs (120 Gb, 1.4 billion nodes, 6.6 billion edges).

Given a real world graph, how should we lay-out its edges? How can we compress it? These questions are closely related, and the typical approach so far is to find clique-like communities, like the cavemen graph', and compress them. We show that the block-diagonal mental image of the cavemen graph' is the wrong paradigm, in full agreement with earlier results that real world graphs have no good cuts. Instead, we propose to envision graphs as a collection of hubs connecting spokes, with super-hubs connecting the hubs, and so on, recursively. Based on the idea, we propose the SLASHBURN method to recursively split a graph into hubs and spokes connected only by the hubs. We also propose techniques to select the hubs and give an ordering to the spokes, in addition to the basic SLASHBURN. We give theoretical analysis of the proposed hub selection methods. Our view point has several advantages: (a) it avoids the no good cuts' problem, (b) it gives better compression, and (c) it leads to faster execution times for matrix-vector operations, which are the back-bone of most graph processing tools. Through experiments, we show that SLASHBURN consistently outperforms other methods for all data sets, resulting in better compression and faster running time. Moreover, we show that SLASHBURN with the appropriate spokes ordering can further improve compression while hardly sacrificing the running time.

Given a set of k networks, possibly with different sizes and no overlaps in nodes or links, how can we quickly assess similarity between them? Analogously, are there a set of social theories which, when represented by a small number of descriptive, numerical features, effectively serve as a "signature" for the network? Having such signatures will enable a wealth of graph mining and social network analysis tasks, including clustering, outlier detection, visualization, etc. We propose a novel, effective, and scalable method, called NETSIMILE, for solving the above problem. Our approach has the following desirable properties: (a) It is supported by a set of social theories. (b) It gives similarity scores that are size-invariant. (c) It is scalable, being linear on the number of links for graph signature extraction. In extensive experiments on numerous synthetic and real networks from disparate domains, NETSIMILE outperforms baseline competitors. We also demonstrate how our approach enables several mining tasks such as clustering, visualization, discontinuity detection, network transfer learning, and re-identification across networks.

Question answering (Q&A) communities have been gaining popularity in the past few years. The success of such sites depends mainly on the contribution of a small number of expert users who provide a significant portion of the helpful answers, and so identifying users that have the potential of becoming strong contributers is an important task for owners of such communities.

How does malware propagate? Does it form spikes over time? Does it resemble the propagation pattern of benign files, such as software patches? Does it spread uniformly over countries? How long does it take for a URL that distributes malware to be detected and shut down?

Clustering is one of the fundamental data mining tasks. While traditional clustering techniques assign each object to a single cluster only, in many applications it has been observed that objects might belong to multiple clusters with different degrees. In this work, we present a Bayesian framework to tackle the challenge of mixed membership clustering for vector data. We exploit the ideas of subspace clustering where the relevance of dimensions might be different for each cluster. Combining the relevance of the dimensions with the cluster membership degree of the objects, we propose a novel type of mixture model able to represent data containing mixed membership subspace clusters. For learning our model, we develop an efficient algorithm based on variational inference allowing easy parallelization. In our empirical study on synthetic and real data we show the strengths of our novel clustering technique.

User review is a crucial component of open mobile app markets such as the Google Play Store. How do we automatically summarize millions of user reviews and make sense out of them? Unfortunately, beyond simple summaries such as histograms of user ratings, there are few analytic tools that can provide insights into user reviews. In this paper, we propose Wiscom, a system that can analyze tens of millions user ratings and comments in mobile app markets at three different levels of detail. Our system is able to (a) discover inconsistencies in reviews; (b) identify reasons why users like or dislike a given app, and provide an interactive, zoomable view of how users' reviews evolve over time; and (c) provide valuable insights into the entire app market, identifying users' major concerns and preferences of different types of apps. Results using our techniques are reported on a 32GB dataset consisting of over 13 million user reviews of 171,493 Android apps in the Google Play Store. We discuss how the techniques presented herein can be deployed to help a mobile app market operator such as Google as well as individual app developers and end-users.

This paper reports on methods and results of an applied research project by a team consisting of SAIC and four universities to develop, integrate, and evaluate new approaches to detect the weak signals characteristic of insider threats on organizations' information systems. Our system combines structural and semantic information from a real corporate database of monitored activity on their users' computers to detect independently developed red team inserts of malicious insider activities. We have developed and applied multiple algorithms for anomaly detection based on suspected scenarios of malicious insider behavior, indicators of unusual activities, high-dimensional statistical patterns, temporal sequences, and normal graph evolution. Algorithms and representations for dynamic graph processing provide the ability to scale as needed for enterprise-level deployments on real-time data streams. We have also developed a visual language for specifying combinations of features, baselines, peer groups, time periods, and algorithms to detect anomalies suggestive of instances of insider threat behavior. We defined over 100 data features in seven categories based on approximately 5.5 million actions per day from approximately 5,500 users. We have achieved area under the ROC curve values of up to 0.979 and lift values of 65 on the top 50 user-days identified on two months of real data.

Preventing contagion in networks is an important problem in public health and other domains. Targeting nodes to immunize based on their network interactions has been shown to be far more effective at stemming infection spread than immunizing random subsets of nodes. However, the assumption that selected nodes can be rendered completely immune does not hold for infections for which there is no vaccination or effective treatment. Instead, one can confer fractional immunity to some nodes by allocating variable amounts of infection-prevention resource to them. We formulate the problem to distribute a fixed amount of resource across nodes in a network such that the infection rate is minimized, prove that it is NP-complete and derive a highly effective and efficient linear-time algorithm. We demonstrate the efficiency and accuracy of our algorithm compared to several other methods using simulation on real-world network datasets including US-MEDICARE and state-level interhospital patient transfer data. We find that concentrating resources at a small subset of nodes using our algorithm is up to 6 times more effective than distributing them uniformly (as is current practice) or using network-based heuristics. To the best of our knowledge, we are the first to formulate the problem, use truly nation-scale network data and propose effective algorithms.

Suppose we are given a large graph in which, by some external process, a handful of nodes are marked. What can we say about these nodes? Are they close together in the graph? or, if segregated, how many groups do they form? We approach this problem by trying to find sets of simple connection pathways between sets of marked nodes. We formalize the problem in terms of the Minimum Description Length principle: a pathway is simple when we need only few bits to tell which edges to follow, such that we visit all nodes in a group. Then, the best partitioning is the one that requires the least number of bits to describe the paths that visit all the marked nodes. We prove that solving this problem is NP-hard, and introduce DOT2DOT, an efficient algorithm for partitioning marked nodes by finding simple pathways between nodes. Experimentation shows that DOT2DOT correctly groups nodes for which good connection paths can be constructed, while separating distant nodes.

How to spot and summarize anomalies in dynamic networks such as road networks, communication networks and social networks? An anomalous event, such as a traffic accident, a denial of service attack or a chemical spill, can affect several near-by edges and make them behave abnormally, over several consecutive time-ticks. We focus on spotting and summarizing such significant anomalous regions, spanning space (i.e. nearby edges), as well as time. Our first contribution is the problem formulation, namely finding all such Significant Anomalous Regions (SAR). The next contribution is the design of novel algorithms: an expensive, exhaustive algorithm, as well as an efficient approximation, called NETSPOT. Compared to the exhaustive algorithm, NETSPOT is up to one order of magnitude faster in real data, while achieving less than 4% average relative error rate. In synthetic datasets, it is more than 30 times faster and solves large problem instances that are otherwise infeasible. The final contribution is the validation on real data: we demonstrate the utility of NETSPOT for inferring accidents on road networks and detecting patterns of anomalous access to subnetworks of Wikipedia. We also study NETSPOT'S scalability in large social, transportation and synthetic evolving networks, spanning in total up to 50 million edges.

How much did a network change since yesterday? How different is the wiring between Bob's brain (a left-handed male) and Alice's brain (a right-handed female)? Graph similarity with known node correspondence, i.e. the detection of changes in the connectivity of graphs, arises in numerous settings. In this work, we formally state the axioms and desired properties of the graph similarity functions, and evaluate when state-of-the-art methods fail to detect crucial connectivity changes in graphs. We propose DeltaCon, a principled, intuitive, and scalable algorithm that assesses the similarity between two graphs on the same nodes (e.g. employees of a company, customers of a mobile carrier). Experiments on various synthetic and real graphs showcase the advantages of our method over existing similarity measures. Finally, we employ DeltaCon to real applications: (a) we classify people to groups of high and low creativity based on their brain connectivity graphs, and (b) do temporal anomaly detection in the who-emails-whom Enron graph.

Detecting anomalies and events in data is a vital task, with numerous applications in security, finance, health care, law enforcement, and many others. While many techniques have been developed in past years for spotting outliers and anomalies in unstructured collections of multi-dimensional points, with graph data becoming ubiquitous, techniques for structured graph data have been of focus recently. As objects in graphs have long-range correlations, novel technology has been developed for abnormality detection in graph data.

How can web services that depend on user generated content discern fraudulent input by spammers from legitimate input? In this paper we focus on the social network Facebook and the problem of discerning ill-gotten Page Likes, made by spammers hoping to turn a profit, from legitimate Page Likes. Our method, which we refer to as CopyCatch, detects lockstep Page Like patterns on Facebook by analyzing only the social graph between users and Pages and the times at which the edges in the graph (the Likes) were created. We offer the following contributions: (1) We give a novel problem formulation, with a simple concrete definition of suspicious behavior in terms of graph structure and edge constraints. (2) We offer two algorithms to find such suspicious lockstep behavior - one provably-convergent iterative algorithm and one approximate, scalable MapReduce implementation. (3) We show that our method severely limits "greedy attacks" and analyze the bounds from the application of the Zarankiewicz problem to our setting. Finally, we demonstrate and discuss the effectiveness of CopyCatch at Facebook and on synthetic data, as well as potential extensions to anomaly detection problems in other domains. CopyCatch is actively in use at Facebook, searching for attacks on Facebook's social graph of over a billion users, many millions of Pages, and billions of Page Likes.

How often do individuals perform a given communication activity in the Web, such as posting comments on blogs or news? Could we have a generative model to create communication events with realistic inter-event time distributions (IEDs)? Which properties should we strive to match? Current literature has seemingly contradictory results for IED: some studies claim good fits with power laws; others with non-homogeneous Poisson processes. Given these two approaches, we ask: which is the correct one? Can we reconcile them all? We show here that, surprisingly, both approaches are correct, being corner cases of the proposed Self-Feeding Process (SFP). We show that the SFP (a) exhibits a unifying power, which generates power law tails (including the so-called "top-concavity" that real data exhibits), as well as short-term Poisson behavior; (b) avoids the "i.i.d. fallacy", which none of the prevailing models have studied before; and (c) is extremely parsimonious, requiring usually only one, and in general, at most two parameters. Experiments conducted on eight large, diverse real datasets (e.g., Youtube and blog comments, e-mails, SMSs, etc) reveal that the SFP mimics their properties very well.

In this paper, we study the intertwined propagation of two competing "memes" (or data, rumors, etc.) in a composite network. Within the constraints of this scenario, we ask two key questions: (a) which meme will prevail? and (b) can one influence the outcome of the propagations? Our model is underpinned by two key concepts, a structural graph model (composite network) and a viral propagation model (SI1I2S). Using this framework, we formulate a non-linear dynamic system and perform an eigenvalue analysis to identify the tipping point of the epidemic behavior. Based on insights gained from this analysis, we demonstrate an effective and accurate prediction method to determine viral dominance, which we call the EigenPredictor. Next, using a combination of synthetic and real composite networks, we evaluate the effectiveness of various viral suppression techniques by either a) concurrently suppressing both memes or b) unilaterally suppressing a single meme while leaving the other relatively unaffected.

This paper proposes Halite, a novel, fast, and scalable clustering method that looks for clusters in subspaces of multidimensional data. Existing methods are typically superlinear in space or execution time. Halite's strengths are that it is fast and scalable, while still giving highly accurate results. Specifically the main contributions of Halite are: 1) Scalability: it is linear or quasi linear in time and space regarding the data size and dimensionality, and the dimensionality of the clusters' subspaces; 2) Usability: it is deterministic, robust to noise, doesn't take the number of clusters as an input parameter, and detects clusters in subspaces generated by original axes or by their linear combinations, including space rotation; 3) Effectiveness: it is accurate, providing results with equal or better quality compared to top related works; and 4) Generality: it includes a soft clustering approach. Experiments on synthetic data ranging from five to 30 axes and up to 1 \rm million points were performed. Halite was in average at least 12 times faster than seven representative works, and always presented highly accurate results. On real data, Halite was at least 11 times faster than others, increasing their accuracy in up to 35 percent. Finally, we report experiments in a real scenario where soft clustering is desirable.

Current applications have produced graphs on the order of hundreds of thousands of nodes and millions of edges. To take advantage of such graphs, one must be able to find patterns, outliers, and communities. These tasks are better performed in an interactive environment, where human expertise can guide the process. For large graphs, though, there are some challenges: the excessive processing requirements are prohibitive, and drawing hundred-thousand nodes results in cluttered images hard to comprehend. To cope with these problems, we propose an innovative framework suited for any kind of tree-like graph visual design. GMine integrates 1) a representation for graphs organized as hierarchies of partitions-the concepts of SuperGraph and Graph-Tree; and 2) a graph summarization methodology-CEPS. Our graph representation deals with the problem of tracing the connection aspects of a graph hierarchy with sub linear complexity, allowing one to grasp the neighborhood of a single node or of a group of nodes in a single click. As a proof of concept, the visual environment of GMine is instantiated as a system in which large graphs can be investigated globally and locally.

We present TOURVIZ, a interactive system for visualizing and making sense of large network datasets. Given a set of user-specified nodes of interest, TOURVIZ integrates 1) novel algorithms to find the best sub graphs that succinctly connect these nodes, and 2) visualization and interaction features that help people explore such sub graphs. We will demonstrate TOURVIZ's usage and benefits using the DBLP co-authorship graph, which consists of 329K authors (nodes) and 1094K co-authorship relations (edges). TOURVIZ can work with any kinds of graphs. We will engage the audience to try our system and comment on its usability, usefulness, and how our system may help with their work and data analytics in their domains.

Spotting anomalies in large multi-dimensional databases is a crucial task with many applications in finance, health care, security, etc. We introduce COMPREX, a new approach for identifying anomalies using pattern-based compression. Informally, our method finds a collection of dictionaries that describe the norm of a database succinctly, and subsequently flags those points dissimilar to the norm---with high compression cost---as anomalies.

The top-n recommendation focuses on finding the top-n items that the target user is likely to purchase rather than predicting his/her ratings on individual items. In this paper, we propose a novel method that provides top-n recommendation by probabilistically determining the target user's preference on items. This method models the purchasing relationships between users and items as a bipartite graph and employs Belief Propagation to compute the preference of the target user on items. We analyze the proposed method in detail by examining the changes in recommendation accuracy under different parameter settings. We also show that the proposed method is up to 40% more accurate than an existing method by comparing it with an RWR-based method via extensive experiments.

Controlling the dissemination of an entity (e.g., meme, virus, etc) on a large graph is an interesting problem in many disciplines. Examples include epidemiology, computer security, marketing, etc. So far, previous studies have mostly focused on removing or inoculating nodes to achieve the desired outcome.

We have entered in an era of big data. Graphs are now measured in terabytes or even petabytes; analyzing them has become increasingly challenging. How do we find patterns and anomalies in these graphs that no longer fit in memory? How should we exploit parallel computation to boost our analysis capabilities? We present PEGASUS, the first open-source, peta-scale graph mining library, for the HADOOP platform (open-source implementation of MAPREDUCE). By observing that many graph mining operations can be described by repeated matrix-vector multiplications, we devised an important primitive called GIM-V for PEGASUS that applies to all such operations. GIM-V (Generalized Iterative Matrix-Vector multiplication) is highly optimized, achieving (1) good scale-up with the number of machines, (2) linear run time on the number of edges, and (3) more than 9 times faster performance over the non-optimized version. We ran experiments for PEGASUS on M45, one of the largest HADOOP clusters in the world. We report our findings on several real graphs with billions of nodes and edges. Selected findings include (a) the discovery of adult advertisers in the who-follows-whom on Twitter, and (b) the 7-degrees of separation in the Web graph.

Given a snapshot of a large graph, in which an infection has been spreading for some time, can we identify those nodes from which the infection started to spread? In other words, can we reliably tell who the culprits are? In this paper we answer this question affirmatively, and give an efficient method called NETSLEUTH for the well-known Susceptible-Infected virus propagation model. Essentially, we are after that set of seed nodes that best explain the given snapshot. We propose to employ the Minimum Description Length principle to identify the best set of seed nodes and virus propagation ripple, as the one by which we can most succinctly describe the infected graph. We give an highly efficient algorithm to identify likely sets of seed nodes given a snapshot. Then, given these seed nodes, we show we can optimize the virus propagation ripple in a principled way by maximizing likelihood. With all three combined, NETSLEUTH can automatically identify the correct number of seed nodes, as well as which nodes are the culprits. Experimentation on our method shows high accuracy in the detection of seed nodes, in addition to the correct automatic identification of their number. Moreover, we show NETSLEUTH scales linearly in the number of nodes of the graph.

Suppose we have two competing ideas/products/viruses, that propagate over a social or other network. Suppose that they are strong/virulent enough, so that each, if left alone, could lead to an epidemic. What will happen when both operate on the network? Earlier models assume that there is perfect competition: if a user buys product 'A' (or gets infected with virus 'X'), she will never buy product 'B' (or virus 'Y'). This is not always true: for example, a user could install and use both Firefox and Google Chrome as browsers. Similarly, one type of flu may give partial immunity against some other similar disease.

We present TourViz, a system that helps its users to interactively visualize and make sense in large network datasets. In particular, it takes as input a set of nodes the user specifies as of interest and presents the user with a visualization of connection subgraphs around these input nodes. Each connection subgraph contains good pathways that highlight succinct connections among a "close-by" group of input nodes. TourViz combines visualization with rich user interaction to engage and help the user to further understand the relations among the nodes of interest,by exploring their neighborhood on demand as well as modifying the set of interest nodes.

Given a network, intuitively two nodes belong to the same role if they have similar structural behavior. Roles should be automatically determined from the data, and could be, for example, "clique-members," "periphery-nodes," etc. Roles enable numerous novel and useful network-mining tasks, such as sense-making, searching for similar nodes, and node classification. This paper addresses the question: Given a graph, how can we automatically discover roles for nodes? We propose RolX (Role eXtraction), a scalable (linear in the number of edges), unsupervised learning approach for automatically extracting structural roles from general network data. We demonstrate the effectiveness of RolX on several network-mining tasks: from exploratory data analysis to network transfer learning. Moreover, we compare network role discovery with network community discovery. We highlight fundamental differences between the two (e.g., roles generalize across disconnected networks, communities do not); and show that the two approaches are complimentary in nature.

Many data are modeled as tensors, or multi dimensional arrays. Examples include the predicates (subject, verb, object) in knowledge bases, hyperlinks and anchor texts in the Web graphs, sensor streams (time, location, and type), social networks over time, and DBLP conference-author-keyword relations. Tensor decomposition is an important data mining tool with various applications including clustering, trend detection, and anomaly detection. However, current tensor decomposition algorithms are not scalable for large tensors with billions of sizes and hundreds millions of nonzeros: the largest tensor in the literature remains thousands of sizes and hundreds thousands of nonzeros.

Given huge collections of time-evolving events such as web-click logs, which consist of multiple attributes (e.g., URL, userID, times- tamp), how do we find patterns and trends? How do we go about capturing daily patterns and forecasting future events? We need two properties: (a) effectiveness, that is, the patterns should help us understand the data, discover groups, and enable forecasting, and (b) scalability, that is, the method should be linear with the data size. We introduce TriMine, which performs three-way mining for all three attributes, namely, URLs, users, and time. Specifically TriMine discovers hidden topics, groups of URLs, and groups of users, simultaneously. Thanks to its concise but effective summarization, it makes it possible to accomplish the most challenging and important task, namely, to forecast future events. Extensive experiments on real datasets demonstrate that TriMine discovers meaningful topics and makes long-range forecasts, which are notoriously difficult to achieve. In fact, TriMine consistently outperforms the best state-of-the-art existing methods in terms of accuracy and execution speed (up to 74x faster).

The recent explosion in the adoption of search engines and new media such as blogs and Twitter have facilitated faster propagation of news and rumors. How quickly does a piece of news spread over these media? How does its popularity diminish over time? Does the rising and falling pattern follow a simple universal law?

Metrics like disk activity and network traffic are widespread sources of diagnosis and monitoring information in datacenters and networks. However, as the scale of these systems increases, examining the raw data yields diminishing insight. We present RainMon, a novel end-to-end approach for mining timeseries monitoring data designed to handle its size and unique characteristics. Our system is able to (a) mine large, bursty, real-world monitoring data, (b) find significant trends and anomalies in the data, (c) compress the raw data effectively, and (d) estimate trends to make forecasts. Furthermore, RainMon integrates the full analysis process from data storage to the user interface to provide accessible long-term diagnosis. We apply RainMon to three real-world datasets from production systems and show its utility in discovering anomalous machines and time periods.

Given a graph with node attributes, how can we find meaningful patterns such as clusters, bridges, and outliers? Attributed graphs appear in real world in the form of social networks with user interests, gene interaction networks with gene expression information, phone call networks with customer demographics, and many others. In effect, we want to group the nodes into clusters with similar connectivity and homogeneous attributes. Most existing graph clustering algorithms either consider only the connectivity structure of the graph and ignore the node attributes, or require several user-defined parameters such as the number of clusters. We propose PICS, a novel, parameter-free method for mining attributed graphs. Two key advantages of our method are that (1) it requires no user-specified parameters such as the number of clusters and similarity functions, and (2) its running time scales linearly with total graph and attribute size. Our experiments show that PICS reveals meaningful and insightful patterns and outliers in both synthetic and real datasets, including call networks, political books, political blogs, and collections from Twitter and YouTube which have more than 70K nodes and 30K attributes.

Given a large social graph, like a scientific collaboration network, what can we say about its robustness? Can we estimate a robustness index for a graph quickly? If the graph evolves over time, how these properties change? In this work, we are trying to answer the above questions studying the expansion properties of large social graphs. First, we present a measure which characterizes the robustness properties of a graph, and serves as global measure of the community structure (or lack thereof). We study how these properties change over time and we show how to spot outliers and anomalies over time. We apply our method on several diverse real networks with millions of nodes. We also show how to compute our measure efficiently by exploiting the special spectral properties of real-world networks.

Given a large graph with millions or billions of nodes and edges, like a who-follows-whom Twitter graph, how do we scalably compute its statistics, summarize its patterns, spot anomalies, visualize and make sense of it? We present OPAvion, a graph mining system that provides a scalable, interactive workflow to accomplish these analysis tasks. OPAvion consists of three modules: (1) The Summarization module (Pegasus) operates off-line on massive, disk-resident graphs and computes graph statistics, like PageRank scores, connected components, degree distribution, triangles, etc.; (2) The Anomaly Detection module (OddBall) uses graph statistics to mine patterns and spot anomalies, such as nodes with many contacts but few interactions with them (possibly telemarketers); (3) The Interactive Visualization module (Apolo) lets users incrementally explore the graph, starting with their chosen nodes or the flagged anomalous nodes; then users can expand to the nodes' vicinities, label them into categories, and thus interactively navigate the interesting parts of the graph.

Graphs are everywhere: social networks, the World Wide Web, biological networks, and many more. The sizes of graphs are growing at unprecedented rate, spanning millions and billions of nodes and edges. What are the patterns in large graphs, spanning Giga, Tera, and heading toward Peta bytes? What are the best tools, and how can they help us solve graph mining problems? How do we scale up algorithms for handling graphs with billions of nodes and edges? These are exactly the goals of this tutorial. We start with the patterns in real-world static, weighted, and dynamic graphs. Then we describe important tools for large graph mining, including singular value decomposition, and Hadoop. Finally, we conclude with the design and the implementation of scalable graph mining algorithms on Hadoop.

Anomaly detection in dynamic networks has a rich gamut of application domains, such as road networks, communication networks and water distribution networks. An anomalous event, such as a traffic accident, denial of service attack or a chemical spill, can cause a local shift from normal behavior in the network state that persists over an interval of time. Detecting such anomalous regions of network and time extent in large real-world networks is a challenging task. Existing anomaly detection techniques focus on either the time series associated with individual network edges or on global anomalies that affect the entire network. In order to detect anomalous regions, one needs to consider both the time and the affected network substructure jointly, which brings forth computational challenges due to the combinatorial nature of possible solutions.

Given two competing products (or memes, or viruses etc.) spreading over a given network, can we predict what will happen at the end, that is, which product will 'win', in terms of highest market share? One may naively expect that the better product (stronger virus) will just have a larger footprint, proportional to the quality ratio of the products (or strength ratio of the viruses). However, we prove the surprising result that, under realistic conditions, for any graph topology, the stronger virus completely wipes-out the weaker one, thus not merely 'winning' but 'taking it all'. In addition to the proofs, we also demonstrate our result with simulations over diverse, real graph topologies, including the social-contact graph of the city of Portland OR (about 31 million edges and 1 million nodes) and internet AS router graphs. Finally, we also provide real data about competing products from Google-Insights, like Facebook-Myspace, and we show again that they agree with our analysis.

This work proposes V-SMART-Join , a scalable MapReduce-based framework for discovering all pairs of similar entities. The V-SMART-Join framework is applicable to sets, multisets, and vectors. V-SMART-Join is motivated by the observed skew in the underlying distributions of Internet traffic, and is a family of 2-stage algorithms, where the first stage computes and joins the partial results, and the second stage computes the similarity exactly for all candidate pairs. The V-SMART-Join algorithms are very efficient and scalable in the number of entities, as well as their cardinalities. They were up to 30 times faster than the state of the art algorithm, VCL , when compared on a real dataset of a small size. We also established the scalability of the proposed algorithms by running them on a dataset of a realistic size, on which VCL never succeeded to finish. Experiments were run using real datasets of IPs and cookies, where each IP is represented as a multiset of cookies, and the goal is to discover similar IPs to identify Internet proxies.

How do contagions spread in population networks? Which group should we market to, for maximizing product penetration? Will a given YouTube video go viral? Who are the best people to vaccinate? What happens when two products compete? The objective of this tutorial is to provide an intuitive and concise overview of most important theoretical results and algorithms to help us understand and manipulate such propagation-style processes on large networks. The tutorial contains three parts: (a) Theoretical results on the behavior of fundamental models; (b) Scalable Algorithms for changing the behavior of these processes e.g., for immunization, marketing etc.; and (c) Empirical Studies of diffusion on blogs and on-line websites like Twitter. The problems we focus on are central in surprisingly diverse areas: from computer science and engineering, epidemiology and public health, product marketing to information dissemination. Our emphasis is on intuition behind each topic, and guidelines for the practitioner.

The multi-million sports-betting market is based on the fact that the task of predicting the outcome of a sports event is very hard. Even with the aid of an uncountable number of descriptive statistics and background information, only a few can correctly guess the outcome of a game or a league. In this work, our approach is to move away from the traditional way of predicting sports events, and instead to model sports leagues as networks of players and teams where the only information available is the work relationships among them. We propose two network-based models to predict the behavior of teams in sports leagues. These models are parameter-free, that is, they do not have a single parameter, and moreover are sport-agnostic: they can be applied directly to any team sports league. First, we view a sports league as a network in evolution, and we infer the implicit feedback behind network changes and properties over the years. Then, we use this knowledge to construct the network-based prediction models, which can, with a significantly high probability, indicate how well a team will perform over a season. We compare our proposed models with other prediction models in two of the most popular sports leagues: the National Basketball Association (NBA) and the Major League Baseball (MLB). Our model shows consistently good results in comparison with the other models and, relying upon the network properties of the teams, we achieved a ≈ 14% rank prediction accuracy improvement over our best competitor.

Modern applications such as web knowledge base, network traffic monitoring and online social networks have made available an unprecedented amount of network data with rich types of interactions carrying multiple attributes, for instance, port number and time tick in the case of network traffic. The design of algorithms to leverage this structured relationship with the power of computing to assist researchers and practitioners for better understanding, exploration and navigation of this space of information has become a challenging, albeit rewarding, topic in social network analysis and data mining. The constantly growing scale and enriching genres of network data always demand higher levels of efficiency, robustness and generalizability where existing approaches with successes on small, homogeneous network data are likely to fall short. We introduce MultiAspectForensics, a handy tool to automatically detect and visualize novel sub graph patterns within a local community of nodes in a heterogenous network, such as a set of vertices that form a dense bipartite graph whose edges share exactly the same set of attributes. We apply the proposed method on three data sets from distinct application domains, present empirical results and discuss insights derived from these patterns discovered. Our algorithm, built on scalable tensor analysis procedures, captures spectral properties of network data and reveals informative signals for subsequent domain-specific study and investigation, such as suspicious port-scanning activities in the scenario of cyber-security monitoring.

Extracting useful knowledge from large network datasets has become a fundamental challenge in many domains, from scientific literature to social networks and the web. We introduce Apolo, a system that uses a mixed-initiative approach - combining visualization, rich user interaction and machine learning - to guide the user to incrementally and interactively explore large network data and make sense of it. Apolo engages the user in bottom-up sensemaking to gradually build up an understanding over time by starting small, rather than starting big and drilling down. Apolo also helps users find relevant information by specifying exemplars, and then using a machine learning method called Belief Propagation to infer which other nodes may be of interest. We evaluated Apolo with twelve participants in a between-subjects study, with the task being to find relevant new papers to update an existing survey paper. Using expert judges, participants using Apolo found significantly more relevant papers. Subjective feedback of Apolo was also very positive.

When a researcher starts with a new topic, it would be very useful if seminal papers in the topic and their relationships are provided in advance. We propose an approach to construct seminal paper genealogy and show the effectiveness and efficiency of our approach.

Color descriptors are one of the important features used in content-based image retrieval. The dominant color descriptor (DCD) represents a few perceptually dominant colors in an image through color quantization. For image retrieval based on DCD, the earth mover's distance and the optimal color composition distance are proposed to measure the dissimilarity between two images. Although providing good retrieval results, both methods are too time-consuming to be used in a large image database. To solve the problem, we propose a new distance function that calculates an approximate earth mover's distance in linear time. To calculate the dissimilarity in linear time, the proposed approach employs the space-filling curve for multidimensional color space. To improve the accuracy, the proposed approach uses multiple curves and adjusts the color positions. As a result, our approach achieves order-of-magnitude time improvement but incurs small errors. We have performed extensive experiments to show the effectiveness and efficiency of the proposed approach. The results reveal that our approach achieves almost the same results with the EMD in linear time.

A blogosphere is a representative example of online social networks. In this paper, we address spectral analysis of a blogosphere. We model a real-world blogosphere as a matrix and a tensor, and then analyze it by using the SVD and PARAFAC decomposition. According to the results, the SVD successfully identified communities, each of which focuses on a specific topic, and also found hub blogs and authoritative posts within each community. The PARAFAC decomposition also succeeded in extracting more communities of finer granules than the SVD. Also, the PARAFAC decomposition could identify the dominant keywords in addition to the hub blogs and authoritative posts honored in each community.

How do we find patterns and anomalies, on graphs with billions of nodes and edges, which do not fit in memory? How to use parallelism for such terabyte-scale graphs? In this work, we focus on inference, which often corresponds, intuitively, to “guilt by association” scenarios. For example, if a person is a drug-abuser, probably its friends are so, too; if a node in a social network is of male gender, his dates are probably females. We show how to do inference on such huge graphs through our proposed HADOOP Line graph Fixed Point (HA-LFP), an efficient parallel algorithm for sparse billion-scale graphs, using the HADOOP platform. Our contributions include (a) the design of HA-LFP, observing that it corresponds to a fixed point on a line graph induced from the original graph; (b) scalability analysis, showing that our algorithm scales up well with the number of edges, as well as with the number of machines; and (c) experimental results on two private, as well as two of the largest publicly available graphs - the Web Graphs from Yahoo! (6.6 billion edges and 0.24 Tera bytes), and the Twitter graph (3.7 billion edges and 0.13 Tera bytes). We evaluated our algorithm using M45, one of the top 50 fastest supercomputers in the world, and we report patterns and anomalies discovered by our algorithm, which would be invisible otherwise.

Given a real world graph, how should we lay-out its edges? How can we compress it? These questions are closely related, and the typical approach so far is to find clique-like communities, like the `cavemen graph', and compress them. We show that the block-diagonal mental image of the `cavemen graph' is the wrong paradigm, in full agreement with earlier results that real world graphs have no good cuts. Instead, we propose to envision graphs as a collection of hubs connecting spokes, with super-hubs connecting the hubs, and so on, recursively. Based on the idea, we propose the Slash Burn method (burn the hubs, and slash the remaining graph into smaller connected components). Our view point has several advantages: (a) it avoids the `no good cuts' problem, (b) it gives better compression, and (c) it leads to faster execution times for matrix-vector operations, which are the back-bone of most graph processing tools. Experimental results show that our Slash Burn method consistently outperforms other methods on all datasets, giving good compression and faster running time.

Given a network of who-contacts-whom or who links-to-whom, will a contagious virus/product/meme spread and 'take-over' (cause an epidemic) or die-out quickly? What will change if nodes have partial, temporary or permanent immunity? The epidemic threshold is the minimum level of virulence to prevent a viral contagion from dying out quickly and determining it is a fundamental question in epidemiology and related areas. Most earlier work focuses either on special types of graphs or on specific epidemiological/cascade models. We are the first to show the G2-threshold (twice generalized) theorem, which nicely de-couples the effect of the topology and the virus model. Our result unifies and includes as special case older results and shows that the threshold depends on the first eigenvalue of the connectivity matrix, (a) for any graph and (b) for all propagation models in standard literature (more than 25, including H.I.V.) [20], [12]. Our discovery has broad implications for the vulnerability of real, complex networks, and numerous applications, including viral marketing, blog dynamics, influence propagation, easy answers to 'what-if' questions, and simplified design and evaluation of immunization policies. We also demonstrate our result using extensive simulations on one of the biggest available social contact graphs containing more than 31 million interactions among more than 1 million people representing the city of Portland, Oregon, USA.

We present APOLO, a system that uses a mixed-initiative approach to help people interactively explore and make sense of large network datasets. It combines visualization, rich user interaction and machine learning to engage the user in bottom-up sensemaking to gradually build up an understanding over time by starting small, rather than starting big and drilling down. APOLO helps users find relevant information by specifying exemplars, and then using a machine learning method called Belief Propagation to infer which other nodes may be of interest. We demonstrate APOLO's usage and benefits using a Google Scholar citation graph, consisting of 83,000 articles (nodes) and 150,000 citations relationships. A demo video of APOLO is available at http://www.cs.cmu.edu/~dchau/apolo/apolo.mp4.

Given a very large moderate-to-high dimensionality dataset, how could one cluster its points? For datasets that don't fit even on a single disk, parallelism is a first class option. In this paper we explore MapReduce for clustering this kind of data. The main questions are (a) how to minimize the I/O cost, taking into account the already existing data partition (e.g., on disks), and (b) how to minimize the network cost among processing nodes. Either of them may be a bottleneck. Thus, we propose the Best of both Worlds -- BoW method, that automatically spots the bottleneck and chooses a good strategy. Our main contributions are: (1) We propose BoW and carefully derive its cost functions, which dynamically choose the best strategy; (2) We show that BoW has numerous desirable features: it can work with most serial clustering methods as a plugged-in clustering subroutine, it balances the cost for disk accesses and network accesses, achieving a very good tradeoff between the two, it uses no user-defined parameters (thanks to our reasonable defaults), it matches the clustering quality of the serial algorithm, and it has near-linear scale-up; and finally, (3) We report experiments on real and synthetic data with billions of points, using up to 1,024 cores in parallel. To the best of our knowledge, our Yahoo! web is the largest real dataset ever reported in the database subspace clustering literature. Spanning 0.2 TB of multi-dimensional data, it took only 8 minutes to be clustered, using 128 cores.

Given a graph, how can we extract good features for the nodes? For example, given two large graphs from the same domain, how can we use information in one to do classification in the other (i.e., perform across-network classification or transfer learning on graphs)? Also, if one of the graphs is anonymized, how can we use information in one to de-anonymize the other? The key step in all such graph mining tasks is to find effective node features. We propose ReFeX (Recursive Feature eXtraction), a novel algorithm, that recursively combines local (node-based) features with neighborhood (egonet-based) features; and outputs regional features -- capturing "behavioral" information. We demonstrate how these powerful regional features can be used in within-network and across-network classification and de-anonymization tasks -- without relying on homophily, or the availability of class labels. The contributions of our work are as follows: (a) ReFeX is scalable and (b) it is effective, capturing regional ("behavioral") information in large graphs. We report experiments on real graphs from various domains with over 1M edges, where ReFeX outperforms its competitors on typical graph mining tasks like network classification and de-anonymization.

Graphs appear in numerous applications including cyber-security, the Internet, social networks, protein networks, recommendation systems, and many more. Graphs with millions or even billions of nodes and edges are common-place. How to store such large graphs efficiently? What are the core operations/queries on those graph? How to answer the graph queries quickly? We propose GBASE, a scalable and general graph management and mining system. The key novelties lie in 1) our storage and compression scheme for a parallel setting and 2) the carefully chosen graph operations and their efficient implementation. We designed and implemented an instance of GBASE using MapReduce/Hadoop. GBASE provides a parallel indexing mechanism for graph mining operations that both saves storage space, as well as accelerates queries. We ran numerous experiments on real graphs, spanning billions of nodes and edges, and we show that our proposed GBASE is indeed fast, scalable and nimble, with significant savings in space and time.

Efficient thermal management is important in modern data centers as cooling consumes up to 50% of the total energy. Unlike previous work, we consider proactive thermal management, whereby servers can predict potential overheating events due to dynamics in data center configuration and workload, giving operators enough time to react. However, such forecasting is very challenging due to data center scales and complexity. Moreover, such a physical system is influenced by cyber effects, including workload scheduling in servers. We propose ThermoCast, a novel thermal forecasting model to predict the temperatures surrounding the servers in a data center, based on continuous streams of temperature and airflow measurements. Our approach is (a) capable of capturing cyberphysical interactions and automatically learning them from data; (b) computationally and physically scalable to data center scales; (c) able to provide online prediction with real-time sensor measurements. The paper's main contributions are: (i) We provide a systematic approach to integrate physical laws and sensor observations in a data center; (ii) We provide an algorithm that uses sensor data to learn the parameters of a data center's cyber-physical system. In turn, this ability enables us to reduce model complexity compared to full-fledged fluid dynamics models, while maintaining forecast accuracy; (iii) Unlike previous simulation-based studies, we perform experiments in a production data center. Using real data traces, we show that ThermoCast forecasts temperature better than a machine learning approach solely driven by data, and can successfully predict thermal alarms 4.2 minutes ahead of time.

We present Polonium, a novel Symantec technology that detects malware through large-scale graph inference. Based on the scalable Belief Propagation algorithm, Polonium infers every file's reputation, flagging files with low reputation as malware. We evaluated Polonium with a billion-node graph constructed from the largest file submissions dataset ever published (60 terabytes). Polonium attained a high true positive rate of 87% in detecting malware; in the field, Polonium lifted the detection rate of existing methods by 10 absolute percentage points. We detail Polonium's design and implementation features instrumental to its success. Polonium has served 120 million people and helped answer more than one trillion queries for file reputation.

How often humans communicate with each other? What are the mechanisms that explain how human actions are distributed over time? Here we answer these questions by studying the time interval between calls and SMS messages in an anonymized, large mobile network, with 3.1 million users, over 200 million phone calls and 300 million SMS messages, spanning 70 GigaBytes. Our first contribution is the Truncated Autocatalytic Process (TAP) model, that explains the time between communication events (ie., times between phone-initiations) for a single individual. The novelty is that the model is ‘autocatalytic’, in the sense that the parameters of the model change, depending on the latest inter-event time: long periods of inactivity in the past result in long periods of inactivity in the future, and vice-versa. We show that the TAP model mimics the inter-event times of the users of our dataset extremely well, despite its parsimony and simplicity. Our second contribution is the TAP-classifier, a classification method based on the inter-event times and in addition to other features. We showed that the inferred sleep intervals and the reciprocity between outgoing and incoming calls are good features to classify users. Finally, analyze the network effects of each class of users and we found surprising results. Moreover, all of our methods are fast, and scale linearly with the number of customers.

Given a large stream of users clicking on web sites, how can we find trends, patterns and anomalies? We have developed a novel method, WindMine, and its fine-tuning sibling, WindMine-part, to find patterns and anomalies in such datasets. Our approach has the following advantages: (a) it is effective in discovering meaningful “building blocks” and patterns such as the lunch-break trend and anomalies, (b) it automatically determines suitable window sizes, and (c) it is fast, with its wall clock time linear on the duration of sequences. Moreover, it can be made sub-quadratic on the number of sequences (WindMine-part), with little loss of accuracy. We examine the effectiveness and scalability by performing experiments on 67 GB of real data (one billion clicks for 30 days). Our proposed WindMine does produce concise, informative and interesting patterns. We also show that WindMine-part can be easily implemented in a parallel or distributed setting, and that, even in a single-machine setting, it can be an order of magnitude faster (up to 70 times) than the plain version.

A blog service company provides a function named BlogCast that exposes quality posts on the blog main page to vitalize a blogosphere. This paper analyzes a new type of information diffusion via BlogCast. We show that there exists a strong halo effect in a blogosphere via thorough investigation on a huge volume of blog data.

What do graphs look like? How do they evolve over time? How to handle a graph with a billion nodes? We present a comprehensive list of static and temporal laws, and some recent observations on real graphs (like, e.g., "eigenSpokes"). For generators, we describe some recent ones, which naturally match all of the known properties of real graphs. Finally, for tools, we present "oddBall" for discovering anomalies and patterns, as well as an overview of the PEGASUS system which is designed for handling Billion-node graphs, running on top of the "hadoop" system.

Given large, multimillion-node graphs (e.g., Facebook, Web-crawls, etc.), how do they evolve over time? How are they connected? What are the central nodes and the outliers? In this article we define the Radius plot of a graph and show how it can answer these questions. However, computing the Radius plot is prohibitively expensive for graphs reaching the planetary scale. There are two major contributions in this article: (a) We propose HADI (HAdoop DIameter and radii estimator), a carefully designed and fine-tuned algorithm to compute the radii and the diameter of massive graphs, that runs on the top of the Hadoop / MapReduce system, with excellent scale-up on the number of available machines (b) We run HADI on several real world datasets including YahooWeb (6B edges, 1/8 of a Terabyte), one of the largest public graphs ever analyzed. Thanks to HADI, we report fascinating patterns on large networks, like the surprisingly small effective diameter, the multimodal/bimodal shape of the Radius plot, and its palindrome motion over time.

We propose the Multi-resolution Correlation Cluster detection (MrCC), a novel, scalable method to detect correlation clusters able to analyze dimensional data in the range of around 5 to 30 axes. Existing methods typically exhibit super-linear behavior in terms of space or execution time. MrCC employs a novel data structure based on multi-resolution and gains over previous approaches in: (a) it finds clusters that stand out in the data in a statistical sense; (b) it is linear on running time and memory usage regarding number of data points and dimensionality of subspaces where clusters exist; (c) it is linear in memory usage and quasi-linear in running time regarding space dimensionality; and (d) it is accurate, deterministic, robust to noise, does not require stating the number of clusters as input parameter, does not perform distance calculation and is able to detect clusters in subspaces generated by original axes or linear combinations of original axes, including space rotation. We performed experiments on synthetic data ranging from 5 to 30 axes and from 12 k to 250 k points, and MrCC outperformed in time five of the recent and related work, being in average 10 times faster than the competitors that also presented high accuracy results for every tested dataset. Regarding real data, MrCC found clusters at least 9 times faster than the competitors, increasing their accuracy in up to 34 percent.

Given a large collection of images, very few of which have labels, how can we guess the labels of the remaining majority, and how can we spot those images that need brand new labels, different from the existing ones? Current automatic labeling techniques usually scale super linearly with the data size, and/or they fail when only a tiny amount of labeled data is provided. In this paper, we propose QMAS (Querying, Mining And Summarization of Multi-modal Databases), a fast solution to the following problems: (i) low-labor labeling (L3) – given a collection of images, very few of which are labeled with keywords, find the most suitable labels for the remaining ones, and (ii) mining and attention routing – in the same setting, find clusters, the top-NO outlier images, and the top-NR representative images. We report experiments on real satellite images, two large sets (1.5GB and 2.25GB) of proprietary images and a smaller set (17MB) of public images. We show that QMAS scales linearly with the data size, being up to 40 times faster than top competitors (GCap), obtaining better or equal accuracy. In contrast to other methods, QMAS does low-labor labeling (L3), that is, it works even with tiny initial label sets. It also solves both presented problems and spots tiles that potentially require new labels.

What do graphs look like? How do they evolve over time? How to handle a graph with a billion nodes? We present a comprehensive list of static and temporal laws, and some recent observations on real graphs (e.g., "eigenSpokes"). For generators, we describe some recent ones, which naturally match all of the known properties of real graphs. Finally, for tools, we present "oddball" for discovering anomalies and patterns, as well as an overview of the PEGASUS system which is designed for handling Billion-node graphs, running on top of the "hadoop" system.

How do connected components evolve? What are the regularities that govern the dynamic growth process and the static snapshot of the connected components? In this work, we study patterns in connected components of large, real-world graphs. First, we study one of the largest static Web graphs with billions of nodes and edges and analyze the regularities among the connected components using GFD(Graph Fractal Dimension) as our main tool. Second, we study several time evolving graphs and find dynamic patterns and rules that govern the dynamics of connected components. We analyze the growth rates of top connected components and study their relation over time. We also study the probability that a newcomer absorbs to disconnected components as a function of the current portion of the disconnected components and the degree of the newcomer. Finally, we propose a generative model that explains both the dynamic growth process and the static regularities of connected components.

Given a large graph, like a computer network, which k nodes should we immunize (or monitor, or remove), to make it as robust as possible against a computer virus attack? We need (a) a measure of the 'Vulnerability' of a given network, (b) a measure of the 'Shield-value' of a specific set of k nodes and (c) a fast algorithm to choose the best such k nodes. We answer all these three questions: we give the justification behind our choices, we show that they agree with intuition as well as recent results in immunology. Moreover, we propose NetShield a fast and scalable algorithm. Finally, we give experiments on large real graphs, where NetShield achieves tremendous speed savings exceeding 7 orders of magnitude, against straightforward competitors.

Advances in data collection and storage capacity have made it increasingly possible to collect highly volatile graph data for analysis. Existing graph analysis techniques are not appropriate for such data, especially in cases where streaming or near-real-time results are required. An example that has drawn significant research interest is the cyber-security domain, where internet communication traces are collected and real-time discovery of events, behaviors, patterns, and anomalies is desired. We propose MetricForensics, a scalable framework for analysis of volatile graphs. MetricForensics combines a multi-level "drill down" approach, a collection of user-selected graph metrics, and a collection of analysis techniques. At each successive level, more sophisticated metrics are computed and the graph is viewed at finer temporal resolutions. In this way, MetricForensics scales to highly volatile graphs by only allocating resources for computationally expensive analysis when an interesting event is discovered at a coarser resolution first. We test MetricForensics on three real-world graphs: an enterprise IP trace, a trace of legitimate and malicious network traffic from a research institution, and the MIT Reality Mining proximity sensor data. Our largest graph has 3M vertices and 32M edges, spanning 4.5 days. The results demonstrate the scalability and capability of MetricForensics in analyzing volatile graphs; and highlight four novel phenomena in such graphs: elbows, broken correlations, prolonged spikes, and lightweight stars.

We introduce a novel Bayesian framework for hybrid community discovery in graphs. Our framework, HCDF (short for Hybrid Community Discovery Framework), can effectively incorporate hints from a number of other community detection algorithms and produce results that outperform the constituent parts. We describe two HCDF-based approaches which are: (1) effective, in terms of link prediction performance and robustness to small perturbations in network structure; (2) consistent, in terms of effectiveness across various application domains; (3) scalable to very large graphs; and (4) nonparametric. Our extensive evaluation on a collection of diverse and large real-world graphs, with millions of links, show that our HCDF-based approaches (a) achieve up to 0.22 improvement in link prediction performance as measured by area under ROC curve (AUC), (b) never have an AUC that drops below 0.91 in the worst case, and (c) find communities that are robust to small perturbations of the network structure as defined by Variation of Information (an entropy-based distance metric).

Given large, multi-million node graphs (e.g., FaceBook, web-crawls, etc.), how do they evolve over time? How are they connected? What are the central nodes and the outliers of the graphs? We show that the Radius Plot (pdf of node radii) can answer these questions. However, computing the Radius Plot is prohibitively expensive for graphs reaching the planetary scale. There are two major contributions in this paper: (a) We propose HADI (HAdoop DIameter and radii estimator), a carefully designed and fine-tuned algorithm to compute the diameter of massive graphs, that runs on the top of the Hadoop/MapReduce system, with excellent scale-up on the number of available machines (b) We run HADI on several real world datasets including YahooWeb (6B edges, 1/8 of a Terabyte), one of the largest public graphs ever analyzed. Thanks to HADI, we report fascinating patterns on large networks, like the surprisingly small effective diameter, the multi-modal/bi-modal shape of the Radius Plot, and its palindrome motion over time.

MOTIVATION: Microarray profiling of mRNA abundance is often ill suited for temporal-spatial analysis of gene expressions in multicellular organisms such as Drosophila. Recent progress in image-based genome-scale profiling of whole-body mRNA patterns via in situ hybridization (ISH) calls for development of accurate and automatic image analysis systems to facilitate efficient mining of complex temporal-spatial mRNA patterns, which will be essential for functional genomics and network inference in higher organisms. RESULTS: We present SPEX(2), an automatic system for embryonic ISH image processing, which can extract, transform, compare, classify and cluster spatial gene expression patterns in Drosophila embryos. Our pipeline for gene expression pattern extraction outputs the precise spatial locations and strengths of the gene expression. We performed experiments on the largest publicly available collection of Drosophila ISH images, and show that our method achieves excellent performance in automatic image annotation, and also finds clusters that are significantly enriched, both for gene ontology functional annotations, and for annotation terms from a controlled vocabulary used by human curators to describe these images. AVAILABILITY: Software will be available at http://www.sailing.cs.cmu.edu/. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

We study the problem of mining and summarizing multiple time series effectively and efficiently. We propose PLiF, a novel method to discover essential characteristics ("fingerprints"), by exploiting the joint dynamics in numerical sequences. Our fingerprinting method has the following benefits: (a) it leads to interpretable features; (b) it is versatile: PLiF enables numerous mining tasks, including clustering, compression, visualization, forecasting, and segmentation, matching top competitors in each task; and (c) it is fast and scalable , with linear complexity on the length of the sequences. We did experiments on both synthetic and real datasets, including human motion capture data (17MB of human motions), sensor data (166 sensors), and network router traffic data (18 million raw updates over 2 years). Despite its generality, PLiF outperforms the top clustering methods on clustering; the top compression methods on compression (3 times better reconstruction error, for the same compression ratio); it gives meaningful visualization and at the same time, enjoys a linear scale-up.

A fundamental challenge in utilizing Web search click data is to infer user-perceived relevance from the search log. Not only is the inference a difficult problem involving statistical reasonings but the bulky size, together with the ever-increasing nature, of the log data imposes extra requirements on scalability. In this paper, we propose the Bayesian Browsing Model (BBM), which performs exact inference of the document relevance, only requires a single pass of the data (i.e., the optimal scalability), and is shown effective. We present two sets of experiments to evaluate the model effectiveness and scalability. On the first set of over 50 million search instances of 1.1 million distinct queries, BBM outperforms the state-of-the-art competitor by 29.2% in log-likelihood while being 57 times faster. On the second click log set, spanning a quarter of petabyte, we showcase the scalability of BBM: we implemented it on a commercial MapReduce cluster, and it took only 3 hours to compute the relevance for 1.15 billion distinct query-URL pairs.

The study of data streams has received considerable attention in various communities (theory, databases, data mining, networking), due to several important applications, such as network analysis, sensor monitoring, financial data analysis, and moving object tracking. Our goal in this article is to monitor multiple numerical streams and determine which pairs are correlated with lags, as well as the value of each such lag. Lag correlations and anticorrelations are frequent and very interesting in practice. For example, a decrease in interest rates typically precedes an increase in house sales by a few months; higher amounts of fluoride in drinking water may lead to fewer dental cavities some years later. Other lag settings include network analysis, sensor monitoring, financial data analysis, and tracking of moving objects. Such data streams are often correlated or anticorrelated, but with unknown lag. We propose BRAID, a method of detecting lag correlations among data streams. BRAID can handle data streams of semi-infinite length incrementally, quickly, and with small resource consumption. However, BRAID requires space and time quadratic on a number of streams k . We also propose ThinBRAID, which is even faster than BRAID, requiring O ( k ) space and time per time tick. Our theoretical analysis shows that BRAID/ThinBRAID can estimate lag correlations with little or, often, with no error. Our experiments on real and realistic data show that BRAID and ThinBRAID detect the correct lag perfectly most of the time (the largest relative error was about 1%), while they are significantly faster (up to 40,000 times) than the naïve implementation.

Triangle counting is an important problem in graph mining. The clustering coefficient and the transitivity ratio,two commonly used measures effectively quantify the triangle density in order to quantify the fact that friends of friends tend to be friends themselves. Furthermore, several successful graph mining applications rely on the number of triangles. In this paper, we study the problem of counting triangles in large, power-law networks. Our algorithm, SparsifyingEigenTriangle, relies on the spectral properties of power-law networks and the Achlioptas-McSherry sparsification process. SparsifyingEigenTriangle is easy to parallelize, fast and accurate. We verify the validity of our approach with several experiments in real-world graphs, where we achieve at the same time high accuracy and important speedup versus a straight-forward exact counting competitor.

Given an author-conference graph, how do we answer proximity queries (e.g., what are the most related conferences for John Smith?); how can we tailor the search result if the user provides additional yes/no type of feedback (e.g., what are the most related conferences for John Smith given that he does not like ICML?)? Given the potential computational complexity, we mainly devote ourselves to addressing the computational issues in this paper by proposing an efficient solution (referred to as iPoG-B) for bipartite graphs. Our experimental results show that the proposed fast solution (iPoGB) achieves significant speedup, while leading to the same ranking result.

In this paper, we describe PEGASUS, an open source peta graph mining library which performs typical graph mining tasks such as computing the diameter of the graph, computing the radius of each node and finding the connected components. as the size of graphs reaches several giga-, tera- or peta-bytes, the necessity for such a library grows too. To the best of our knowledge, PEGASUS is the first such library, implemented on the top of the HADOOP platform, the open source version of MAPREDUCE. Many graph mining operations (PageRank, spectral clustering, diameter estimation, connected components etc.) are essentially a repeated matrix-vector multiplication. In this paper we describe a very important primitive for PEGASUS, called GIM-V (generalized iterated matrix-vector multiplication). GIM-V is highly optimized, achieving (a) good scale-up on the number of available machines (b) linear running time on the number of edges, and (c) more than 5 times faster performance over the non-optimized version of GIM-V. Our experiments ran on M45, one of the top 50 supercomputers in the world. We report our findings on several real graphs, including one of the largest publicly available Web graphs, thanks to Yahoo!, with ¿ 6,7 billion edges.

Given a real, and weighted person-to-person network which changes over time, what can we say about the cliques that it contains? Do the incidents of communication, or weights on the edges of a clique follow any pattern? Real, and in-person social networks have many more triangles than chance would dictate. As it turns out, there are many more cliques than one would expect, in surprising patterns.

Given multiple time sequences with missing values, we propose DynaMMo which summarizes, compresses, and finds latent variables. The idea is to discover hidden variables and learn their dynamics, making our algorithm able to function even when there are missing values.We performed experiments on both real and synthetic datasets spanning several megabytes, including motion capture sequences and chlorine levels in drinking water. We show that our proposed DynaMMo method (a) can successfully learn the latent variables and their evolution; (b) can provide high compression for little loss of reconstruction accuracy; (c) can extract compact but powerful features for segmentation, interpretation, and forecasting; (d) has complexity linear on the duration of sequences.

Given a quarter of petabyte click log data, how can we estimate the relevance of each URL for a given query? In this paper, we propose the Bayesian Browsing Model (BBM), a new modeling technique with following advantages: (a) it does exact inference; (b) it is single-pass and parallelizable; (c) it is effective.

Classifying nodes in networks is a task with a wide range of applications. It can be particularly useful in anomaly and fraud detection. Many resources are invested in the task of fraud detection due to the high cost of fraud, and being able to automatically detect potential fraud quickly and precisely allows human investigators to work more efficiently. Many data analytic schemes have been put into use; however, schemes that bolster link analysis prove promising. This work builds upon the belief propagation algorithm for use in detecting collusion and other fraud schemes. We propose an algorithm called SNARE (Social Network Analysis for Risk Evaluation). By allowing one to use domain knowledge as well as link knowledge, the method was very successful for pinpointing misstated accounts in our sample of general ledger data, with a significant improvement over the default heuristic in true positive rates, and a lift factor of up to 6.5 (more than twice that of the default heuristic). We also apply SNARE to the task of graph labeling in general on publicly-available datasets. We show that with only some information about the nodes themselves in a network, we get surprisingly high accuracy of labels. Not only is SNARE applicable in a wide variety of domains, but it is also robust to the choice of parameters and highly scalable-linearly with the number of edges in a graph.

Most of recommender systems try to find items that are most relevant to the older choices of a given user. Here we focus on the "surprise me" query: A user may be bored with his/her usual genre of items (e.g., books, movies, hobbies), and may want a recommendation that is related, but off the beaten path, possibly leading to a new genre of books/movies/hobbies.

The Border Gateway Protocol (BGP) is one of the fundamental computer communication protocols. Monitoring and mining BGP update messages can directly reveal the health and stability of Internet routing. Here we make two contributions: firstly we find patterns in BGP updates, like self-similarity, power-law and lognormal marginals; secondly using these patterns, we find anomalies. Specifically, we develop BGP-lens, an automated BGP updates analysis tool, that has three desirable properties: (a) It is effective, able to identify phenomena that would otherwise go unnoticed, such as a peculiar 'clothesline' behavior or prolonged 'spikes' that last as long as 8 hours; (b) It is scalable, using algorithms are all linear on the number of time-ticks; and (c) It is admin-friendly, giving useful leads for phenomenon of interest.

Counting the number of triangles in a graph is a beautiful algorithmic problem which has gained importance over the last years due to its significant role in complex network analysis. Metrics frequently computed such as the clustering coefficient and the transitivity ratio involve the execution of a triangle counting algorithm. Furthermore, several interesting graph mining applications rely on computing the number of triangles in the graph of interest.

Graphs appear in several settings, like social networks, recommendation systems, computer communication networks, gene/protein biological networks, among others. A deep, recurring question is “What do real graphs look like?” That is, how can we separate real ones from synthetic or real graphs with masked portions? The main contribution of this paper is ShatterPlots, a simple and powerful algorithm to extract patterns from real graphs that help us spot fake/masked graphs. The idea is to shatter a graph, by deleting edges, force it to reach a critical (“Shattering”) point, and study the properties at that point. One of the most striking patterns is the “30-per-cent“: at the Shattering point, all real and synthetic graphs have about 30% more nodes than edges. One of our most discriminative patterns is the “NodeShatteringRatio“, which can almost perfectly separate the real graphs from the synthetic ones of our extensive collection. Additional contributions of this paper are (a) the careful, scalable design of the algorithm, which requires only O(E) time, (b) extensive experiments in a large collection of graphs (19 in total), with up to hundreds of thousands of nodes and million edges, and (c) a wealth of observations and patterns, which show how to distinguish synthetic or masked graphs from real ones.

Given a terabyte click log, can we build an efficient and effective click model? It is commonly believed that web search click logs are a gold mine for search business, because they reflect users' preference over web documents presented by the search engine. Click models provide a principled approach to inferring user-perceived relevance of web documents, which can be leveraged in numerous applications in search businesses. Due to the huge volume of click data, scalability is a must.We present the click chain model (CCM), which is based on a solid, Bayesian framework. It is both scalable and incremental, perfectly meeting the computational challenges imposed by the voluminous click logs that constantly grow. We conduct an extensive experimental study on a data set containing 8.8 million query sessions obtained in July 2008 from a commercial search engine. CCM consistently outperforms two state-of-the-art competitors in a number of metrics, with over 9.7% better log-likelihood, over 6.2% better click perplexity and much more robust (up to 30%) prediction of the first and the last clicked position.

A similarity query considers an element as the query center and searches a dataset to find either the elements far up to a bounding radius or the k nearest ones from the query center. Several algorithms have been developed to efficiently execute similarity queries. However, there are queries that require more than one center, which we call Aggregate Similarity Queries. Such queries appear when the user gives multiple desirable examples, and requests data elements that are similar to all of the examples, as in the case of applying relevance feedback. Here we give the first algorithms that can handle aggregate similarity queries on Metric Access Methods (MAM) such as the M-tree and Slim-tree. Our method, which we call Metric Aggregate Similarity Search (MASS) has the following properties: (a) it requires only the triangle inequality property; (b) it guarantees no false-dismissals, as we prove that it lower-bounds the aggregate distance scores; (c) it can work with any MAM; (d) it can handle any number of query centers, which are either scattered all over the space or concentrated on a restricted region. Experiments on both real and synthetic data show that our method scales on both the number of elements and, if the dataset is in a spatial domain, also on its dimensionality. Moreover, it achieves better results than previous related methods.

Given a collection of complex, time-stamped events, how do we find patterns and anomalies? Events could be meetings with one or more persons and one or more agenda items at zero or more locations (e.g., teleconferences), or they could be publications with authors, keywords, publishers, etc. In such settings, we want to find time stamps that look similar to each other and group them; we also want to find anomalies. In addition, we want our approach to provide interpretations of the clusters and anomalies by annotating them. Furthermore, we want our approach to automatically find the right time-granularity in which to do analysis. Lastly, we want fast, scalable algorithms for all these problems.

We propose an approach for learning visual models of object categories in an unsupervised manner in which we first build a large-scale complex network which captures the interactions of all unit visual features across the entire training set and we infer information, such as which features are in which categories, directly from the graph by using link analysis techniques. The link analysis techniques are based on well-established graph mining techniques used in diverse applications such as WWW, bioinformatics, and social networks. The techniques operate directly on the patterns of connections between features in the graph rather than on statistical properties, e.g., from clustering in feature space. We argue that the resulting techniques are simpler, and we show that they perform similarly or better compared to state of the art techniques on common data sets. We also show results on more challenging data sets than those that have been used in prior work on unsupervised modeling.

Similarity joins have attracted significant interest, with applications in geographical information systems, astronomy, marketing analyzes, and anomaly detection. However, all the past algorithms, although highly fine-tuned, suffer an output explosion if the query range is even moderately large relative to the local data density. Under such circumstances, the response time and the search effort are both almost quadratic in the database size, which is often prohibitive. We solve this problem by providing two algorithms that find a compact representation of the similarity join result, while retaining all the information in the standard join. Our algorithms have the following characteristics: (a) they are at least as fast as the standard similarity join algorithm, and typically much faster, (b) they generate significantly smaller output, (c) they provably lose no information, (d) they scale well to large data sets, and (e) they can be applied to any of the standard tree data structures. Experiments on real and realistic point-sets show that our algorithms are up to several orders of magnitude faster.

Given publication titles and authors, what can we say about the evolution of scientific topics and communities over time? Which communities shrunk, which emerged, and which split, over time? And, when in time were the turning points? We propose TimeFall, which can automatically answer these questions given a social network/graph that evolves over time. The main novelty of the proposed approach is that it needs no user-defined parameters, relying instead on the principle of minimum description length (MDL), to extract the communities, and to find good cut-points in time when communities change abruptly: a cut-point is good, if it leads to shorter data description. We illustrate our algorithm on synthetic and large real datasets, and we show that the results of the TimeFall agree with human intuition.

How do real, weighted graphs change over time? What patterns, if any, do they obey? Earlier studies focus on unweighted graphs, and, with few exceptions, they focus on static snapshots. Here, we report patterns we discover on several real, weighted, time-evolving graphs. The reported patterns can help in detecting anomalies in natural graphs, in making link prediction and in providing more criteria for evaluation of synthetic graph generators. We further propose an intuitive and easy way to construct weighted, time-evolving graphs. In fact, we prove that our generator will produce graphs which obey many patterns and laws observed to date. We also provide empirical evidence to support our claims.

Vector data are normally used for probabilistic graphical models with Bayesian inference. However, tensor data, i.e., multidimensional arrays, are actually natural representations of a large amount of real data, in data mining, computer vision, and many other applications. Aiming at breaking the huge gap between vectors and tensors in conventional statistical tasks, e.g., automatic model selection, this paper proposes a decoupled probabilistic algorithm, named Bayesian tensor analysis (BTA). BTA automatically selects a suitable model for tensor data, as demonstrated by empirical studies.

How do graphs look like? How do they evolve over time? How can we generate realistic-looking graphs? We review some static and temporal 'laws,' and we describe the "Kronecker" graph generator, which naturally matches all of the known properties of real graphs. Moreover, we present tools for discovering anomalies and patterns in two types of graphs, static and time-evolving. For the former, we present the 'CenterPiece' subgraphs (CePS), which expects q query nodes (eg., suspicious and finds the node that is best connected to all q of them (eg., the master mind of a criminal group). We also show how to compute CenterPiece subgraphs efficiently. For the time evolving graphs, we present tensor-based methods, and apply them on real data, like the DBLP author-paper dataset, where they are able to find natural research communities, and track their evolution. Finally, we also briefly mention some results on influence and virus propagation on real graphs, as well as on the emerging map/reduce approach and its impact on large graph mining.

MOTIVATION: Protein complexes integrate multiple gene products to coordinate many biological functions. Given a graph representing pairwise protein interaction data one can search for subgraphs representing protein complexes. Previous methods for performing such search relied on the assumption that complexes form a clique in that graph. While this assumption is true for some complexes, it does not hold for many others. New algorithms are required in order to recover complexes with other types of topological structure. RESULTS: We present an algorithm for inferring protein complexes from weighted interaction graphs. By using graph topological patterns and biological properties as features, we model each complex subgraph by a probabilistic Bayesian network (BN). We use a training set of known complexes to learn the parameters of this BN model. The log-likelihood ratio derived from the BN is then used to score subgraphs in the protein interaction graph and identify new complexes. We applied our method to protein interaction data in yeast. As we show our algorithm achieved a considerable improvement over clique based algorithms in terms of its ability to recover known complexes. We discuss some of the new complexes predicted by our algorithm and determine that they likely represent true complexes. AVAILABILITY: Matlab implementation is available on the supporting website: www.cs.cmu.edu/~qyj/SuperComplex.

We address the problem of classification in partially labeled networks (a.k.a. within-network classification) where observed class labels are sparse. Techniques for statistical relational learning have been shown to perform well on network classification tasks by exploiting dependencies between class labels of neighboring nodes. However, relational classifiers can fail when unlabeled nodes have too few labeled neighbors to support learning (during training phase) and/or inference (during testing phase). This situation arises in real-world problems when observed labels are sparse.

By now, online social networks have become an indispensable part of both online and offline lives of human beings. A large fraction of time spent online by a user is directly influence by the social networks to which he/she belongs. This calls for a deeper examination of social networks as large-scale dynamic objects that foster efficient person-person interaction.

Multi-core processors with ever increasing number of cores per chip are becoming prevalent in modern parallel computing. Our goal is to make use of the multi-core as well as multi-processor architectures to speed up data mining algorithms. Specifically, we present a parallel algorithm for approximate learning of Linear Dynamical Systems (LDS), also known as Kalman Filters (KF). LDSs are widely used in time series analysis such as motion capture modeling, visual tracking etc. We propose Cut-And-Stitch (CAS), a novel method to handle the data dependencies from the chain structure of hidden variables in LDS, so as to parallelize the EM-based parameter learning algorithm. We implement the algorithm using OpenMP on both a supercomputer and a quad-core commercial desktop. The experimental results show that parallel algorithms using Cut-And-Stitch achieve comparable accuracy and almost linear speedups over the serial version. In addition, Cut-And-Stitch can be generalized to other models with similar linear structures such as Hidden Markov Models (HMM) and Switching Kalman Filters (SKF).

The vast majority of earlier work has focused on graphs which are both connected (typically by ignoring all but the giant connected component), and unweighted. Here we study numerous, real, weighted graphs, and report surprising discoveries on the way in which new nodes join and form links in a social network. The motivating questions were the following: How do connected components in a graph form and change over time? What happens after new nodes join a network -- how common are repeated edges? We study numerous diverse, real graphs (citation networks, networks in social media, internet traffic, and others); and make the following contributions: (a) we observe that the non-giant connected components seem to stabilize in size, (b) we observe the weights on the edges follow several power laws with surprising exponents, and (c) we propose an intuitive, generative model for graph growth that obeys observed patterns.

We analyze a massive social network, gathered from the records of a large mobile phone operator, with more than a million users and tens of millions of calls. We examine the distributions of the number of phone calls per customer; the total talk minutes per customer; and the distinct number of calling partners per customer. We find that these distributions are skewed, and that they significantly deviate from what would be expected by power-law and lognormal distributions.

Low-rank approximations of the adjacency matrix of a graph are essential in finding patterns (such as communities) and detecting anomalies. Additionally, it is desirable to track the low-rank structure as the graph evolves over time, efficiently and within limited storage. Real graphs typically have thousands or millions of nodes, but are usually very sparse. However, standard decompositions such as SVD do not preserve sparsity. This has led to the development of methods such as CUR and CMD, which seek a non-orthogonal basis by sampling the columns and/or rows of the sparse matrix.

Semi-supervised learning plays an important role in the recent literature on machine learning and data mining and the developed semisupervised learning techniques have led to many data mining applications in recent years. This paper addresses the semi-supervised learning problem by developing a semiparametric regularization based approach, which attempts to discover the marginal distribution of the data to learn the parametric function through exploiting the geometric distribution of the data. This learned parametric function can then be incorporated into the supervised learning on the available labeled data as the prior knowledge. Specifically, our contributions are: (1) We present a semi-supervised learning approach which incorporates the unlabeled data into the supervised learning by a parametric function learned from the whole data including the labeled and unlabeled data. The parametric function reflects the geometric structure of the marginal distribution of the data. Furthermore, the proposed approach which naturally extends to the out-of-sample data is an inductive learning method in nature. (2) This approach allows a family of algorithms to be developed based on various choices of the original RKHS and the loss function. (3) We provide experimental comparisons showing that the proposed approach leads the state-of-the-art performance on a variety of classification tasks. In particular, we demonstrate that this approach can be used successfully in both transductive and semisupervised settings. 1

We define and solve the problem of “distribution classification”, and, in general, “distribution mining”. Given n distributions (i.e., clouds) of multi-dimensional points, we want to classify them into k classes, to find patterns, rules and out-lier clouds. For example, consider the 2-d case of sales of items, where, for each item sold, we record the unit price and quantity; then, each customer is represented as a distribution/cloud of 2-d points (one for each item he bought). We want to group similar users together, e.g., for market segmentation, anomaly/fraud detection. We propose D-Mine to achieve this goal. Our main contribution is Theorem 3.1, which shows how to use wavelets to speed up the cloud-similarity computations. Extensive experiments on both synthetic and real multi-dimensional data sets show that our method achieves up to 400 faster wall-clock time over the naive implementation, with comparable (and occasionally better) classification quality.

Given an author-conference network that evolves over time, which are the conferences that a given author is most closely related with, and how do they change over time? Large time-evolving bipartite graphs appear in many settings, such as social networks, co-citations, market-basket analysis, and collaborative filtering. Our goal is to monitor (i) the centrality of an individual node (e.g., who are the most important authors?); and (ii) the proximity of two nodes or sets of nodes (e.g., who are the most important authors with respect to a particular conference?) Moreover, we want to do this efficiently and incrementally, and to provide “any-time” answers. We propose pTrack and cTrack, which are based on random walk with restart, and use powerful matrix tools. Experiments on real data show that our methods are effective and efficient: the mining results agree with intuition; and we achieve up to 15∼176 times speed-up, without any quality loss.

How can we efficiently find a clustering, i.e. a concise description of the cluster structure, of a given data set which contains an unknown number of clusters of different shape and distribution and is contaminated by noise? Most existing clustering methods are restricted to the Gaussian cluster model and are very sensitive to noise. If the cluster content follows a non-Gaussian distribution and/or the data set contains a few outliers belonging to no cluster, then the computed data distribution does not match well the true data distribution, or an unnaturally high number of clusters is required to represent the true data distribution of the data set. In this paper we propose OCI (Outlier-robust Clustering using Independent Components), a clustering method which overcomes these problems by (1) applying the exponential power distribution (EPD) as cluster model which is a generalization of Gaussian, uniform, Laplacian and many other distribution functions, (2) applying the Independent Component Analysis (ICA) for both determining the main directions inside a cluster as well as finding split planes in a top-down clustering approach, and (3) defining an efficient and effective filter for outliers, based on EPD and ICA. Our method is parameter-free and as a top-down clustering approach very efficient. An extensive experimental evaluation shows both the accuracy of the obtained clustering result as well as the efficiency of our method.

The amount of biological data publicly available has experienced an exponential growth as the technology advances. Online databases are now playing an important role as information repositories as well as easily accessible platforms for researchers to communicate and contribute. Recent research projects in image bioinformatics produce a number of databases of images, which visualize the spatial expression pattern of a gene (eg. "fj"), and most of which also have one or several annotation keywords (eg., "embryonic hindgut"). C-DEM is an online system for Drosophila (= fruit-fly) Embryo images Mining. It supports queries from all three modalities to all three, namely, (a) genes, (b) images of gene expression, and (c) annotation keywords of the images. Thus, it can find images that are similar to a given image, and/or related to the desirable annotation keywords, and/or related to specific genes. Typical queries are what are most suitable keywords to assign to image insitu28465.jpg or find images that are related to gene "fj", and to the keyword "embryonic hindgut" . C-DEM uses state-of-the-art feature extraction methods for images (wavelets and principal component analysis). It envisions the whole database as a tri-partite graph (one type for each modality), and it uses fast and flexible proximity measures, namely, random walk with restarts (RWR). In addition to flexible querying, C-DEM allows for navigation: the user can click on the results of an earlier query (image thumbnails and/or keywords and/or genes), and the system will report the most related images (and keywords, and genes). The demo is on a real Drosophila Embryo database, with 10,204 images, 2,969 distinct genes, and 113 annotation keywords. The query response time is below one second on a commodity desktop.

Effectively modeling a collection of three-dimensional (3-D) faces is an important task in various applications, especially facial expression-driven ones, e.g., expression generation, retargeting, and synthesis. These 3-D faces naturally form a set of second-order tensors-one modality for identity and the other for expression. The number of these second-order tensors is three times of that of the vertices for 3-D face modeling. As for algorithms, Bayesian data modeling, which is a natural data analysis tool, has been widely applied with great success; however, it works only for vector data. Therefore, there is a gap between tensor-based representation and vector-based data analysis tools. Aiming at bridging this gap and generalizing conventional statistical tools over tensors, this paper proposes a decoupled probabilistic algorithm, which is named Bayesian tensor analysis (BTA). Theoretically, BTA can automatically and suitably determine dimensionality for different modalities of tensor data. With BTA, a collection of 3-D faces can be well modeled. Empirical studies on expression retargeting also justify the advantages of BTA.

How will a virus propagate in a real network? How long does it take to disinfect a network given particular values of infection rate and virus death rate? What is the single best node to immunize? Answering these questions is essential for devising network-wide strategies to counter viruses. In addition, viral propagation is very similar in principle to the spread of rumors, information, and “fads,” implying that the solutions for viral propagation would also offer insights into these other problem settings. We answer these questions by developing a nonlinear dynamical system (NLDS) that accurately models viral propagation in anyarbitrarynetwork, including real and synthesized network graphs. We propose a general epidemic threshold condition for theNLDSsystem: we prove that the epidemic threshold for a network is exactly the inverse of the largest eigenvalue of its adjacency matrix. Finally, we show that below the epidemic threshold, infections die out at an exponential rate. Our epidemic threshold model subsumes many known thresholds for special-case graphs (e.g., Erdös--Rényi, BA powerlaw, homogeneous). We demonstrate the predictive power of our model with extensive experiments on real and synthesized graphs, and show that our threshold condition holds for arbitrary graphs. Finally, we show how to utilize our threshold condition for practical uses: It can dictate which nodes to immunize; it can assess the effects of a throttling policy; it can help us design network topologies so that they are more resistant to viruses.

How do we find patterns in author-keyword associations, evolving over time? Or in data cubes (tensors), with product-branchcustomer sales information? And more generally, how to summarize high-order data cubes (tensors)? How to incrementally update these patterns over time? Matrix decompositions, like principal component analysis (PCA) and variants, are invaluable tools for mining, dimensionality reduction, feature selection, rule identification in numerous settings like streaming data, text, graphs, social networks, and many more settings. However, they have only two orders (i.e., matrices, like author and keyword in the previous example). We propose to envision such higher-order data as tensors, and tap the vast literature on the topic. However, these methods do not necessarily scale up, let alone operate on semi-infinite streams. Thus, we introduce a general framework, incremental tensor analysis (ITA), which efficiently computes a compact summary for high-order and high-dimensional data, and also reveals the hidden correlations. Three variants of ITA are presented: (1) dynamic tensor analysis (DTA); (2) streaming tensor analysis (STA); and (3) window-based tensor analysis (WTA). In paricular, we explore several fundamental design trade-offs such as space efficiency, computational cost, approximation accuracy, time dependency, and model complexity. We implement all our methods and apply them in several real settings, such as network anomaly detection, multiway latent semantic indexing on citation networks, and correlation study on sensor measurements. Our empirical studies show that the proposed methods are fast and accurate and that they find interesting patterns and outliers on the real datasets.

The goal of this paper is to monitor numerical streams, and to find subsequences that are similar to a given query sequence, under the DTW (dynamic time warping) distance. Applications include word spotting, sensor pattern matching, and monitoring of bio-medical signals (e.g., EKG, ECG), and monitoring of environmental (seismic and volcanic) signals. DTW is a very popular distance measure, permitting accelerations and decelerations, and it has been studied for finite, stored sequence sets. However, in many applications such as network analysis and sensor monitoring, massive amounts of data arrive continuously and it is infeasible to save all the historical data. We propose SPRING, a novel algorithm that can solve the problem. We provide a theoretical analysis and prove that SPRING does not sacrifice accuracy, while it requires constant space and time per time-tick. These are dramatic improvements over the naive method. Our experiments on real and realistic data illustrate that SPRING does indeed detect the qualifying subsequences correctly and that it can offer dramatic improvements in speed over the naive implementation.

This paper presents a max margin framework on image annotation and multimodal image retrieval as a structured prediction model. Following the max margin approach the image retrieval problem is formulated as a quadratic programming problem. By properly selecting joint feature representation between different modalities, our framework captures the dependency information between different modalities and avoids retraining the model from scratch when database undergoes dynamic updates. While this framework is a general approach which can be applied to multimodal information retrieval in any domains, we apply this approach to the Berkeley Drosophila embryo image database for the evaluation purpose. Experimental results show significant performance improvements over a state-of-the-art method.

Given a large, real graph, how can we generate a synthetic graph that matches its properties, i.e., it has similar degree distribution, similar (small) diameter, similar spectrum, etc? We propose to use "Kronecker graphs", which naturally obey all of the above properties, and we present KronFit, a fast and scalable algorithm for fitting the Kronecker graph generation model to real networks. A naive approach to fitting would take super-exponential time. In contrast, KronFit takes linear time, by exploiting the structure of Kronecker product and by using sampling. Experiments on large real and synthetic graphs show that KronFit indeed mimics very well the patterns found in the target graphs. Once fitted, the model parameters and the resulting synthetic graphs can be used for anonymization, extrapolations, and graph summarization.

Consider a network of, say, sensors, or P2P nodes, or Bluetooth-enabled cell-phones, where nodes transmit information to each other and where links and nodes can go up or down. Consider also a 'datum', that is, a piece of information, like a report of an emergency condition in a sensor network, a national traditional song, or a mobile phone virus. How often should nodes transmit the datum to each other, so that the datum can survive (or, in the virus case, under what conditions will the virus die out)? Clearly, the link and node fault probabilities are important - what else is needed to ascertain the survivability of the datum? We propose and solve the problem using non-linear dynamical systems and fixed point stability theorems. We provide a closed-form formula that, surprisingly, depends on only one additional parameter, the largest eigenvalue of the connectivity matrix. We illustrate the accuracy of our analysis on realistic and real settings, like mote sensor networks from Intel and MIT, as well as Gnutella and P2P networks.

The problem of multimodal data mining in a multimedia database can be addressed as a structured prediction problem where we learn the mapping from an input to the structured and interdependent output variables. In this paper, built upon the existing literature on the max margin based learning, we develop a new max margin learning approach called Enhanced Max Margin Learning (EMML) framework. In addition, we apply EMML framework to developing an effective and efficient solution to the multimodal data mining problem in a multimedia database. The main contributions include: (1) we have developed a new max margin learning approach - the enhanced max margin learning framework that is much more efficient in learning with a much faster convergence rate, which is verified in empirical evaluations; (2) we have applied this EMML approach to developing an effective and efficient solution to the multimodal data mining problem that is highly scalable in the sense that the query response time is independent of the database scale, allowing facilitating a multimodal data mining querying to a very large scale multimedia database,and excelling many existing multimodal data mining methods in the literature that do not scale up at all; this advantage is also supported through the complexity analysis as well as empirical evaluations against a state-of-the-art multimodal data mining method from the literature. While EMML is a general framework, for the evaluation purpose, we apply it to the Berkeley Drosophila embryo image database, and report the performance comparison with a state-of-the-art multimodal data mining method.

Given a water distribution network, where should we place sensors toquickly detect contaminants? Or, which blogs should we read to avoid missing important stories?.

How can we find communities in dynamic networks of socialinteractions, such as who calls whom, who emails whom, or who sells to whom? How can we spot discontinuity time-points in such streams of graphs, in an on-line, any-time fashion? We propose GraphScope, that addresses both problems, using information theoretic principles. Contrary to the majority of earlier methods, it needs no user-defined parameters. Moreover, it is designed to operate on large graphs, in a streaming fashion. We demonstrate the efficiency and effectiveness of our GraphScope on real datasets from several diverse domains. In all cases it produces meaningful time-evolving patterns that agree with human intuition.

We focus on large graphs where nodes have attributes, such as a social network where the nodes are labelled with each person's job title. In such a setting, we want to find subgraphs that match a user query pattern. For example, a "star" query would be, "find a CEO who has strong interactions with a Manager, a Lawyer,and an Accountant, or another structure as close to that as possible". Similarly, a "loop" query could help spot a money laundering ring.

In this paper we study asymmetric proximity measures on directed graphs, which quantify the relationships between two nodes or two groups of nodes. The measures are useful in several graph mining tasks, including clustering, link prediction and connection subgraph discovery. Our proximity measure is based on the conceptof escape probability. This way, we strive to summarize the multiple facets of nodes-proximity, while avoiding some of the pitfalls to which alternative proximity measures are susceptible. A unique feature of the measures is accounting for the underlying directional information. We put a special emphasis on computational efficiency, and develop fast solutions that are applicable in several settings. Our experimental study shows the usefulness of our proposed direction-aware proximity method for several applications, and that our algorithms achieve a significant speedup (up to 50,000x) over straight forward implementations.

How do blogs cite and influence each other? How do such links evolve? Does the popularity of old blog posts drop exponentially with time? These are some of the questions that we address in this work. Blogs (weblogs) have become an important medium of information because of their timely publication, ease of use, and wide availability. In fact, they often make headlines, by discussing and discovering evidence about political events and facts. Often blogs link to one another, creating a publicly available record of how information and influence spreads through an underlying social network. Aggregating links from several blog posts creates a directed graph which we analyze to discover the patterns of information propagation in blogspace, and thereby understand the underlying social network. Here we report some surprising findings of the blog linking and information propagation structure, after we analyzed one of the largest available datasets, with 45,000 blogs and ≈ 2.2 million blog-postings. Our analysis also sheds light on how rumors, viruses, and ideas propagate over social and computer networks.

Given a large sparse graph, how can we find patterns and anomalies? Several important applications can be modeled as large sparse graphs, e.g., network traffic monitoring, research citation network analysis, social network analysis, and regulatory networks in genes. Low rank decompositions, such as SVD and CUR, are powerful techniques for revealing latent/hidden variables and associated patterns from high dimensional data. However, those methods often ignore the sparsity property of the graph, and hence usually incur too high memory and computational cost to be practical. We propose a novel method, the Compact Matrix Decomposition (CMD), to compute sparse low rank approximations. CMD dramatically reduces both the computation cost and the space requirements over existing decomposition methods (SVD, CUR). Using CMD as the key building block, we further propose procedures to efficiently construct and analyze dynamic graphs from real-time application data. We provide theoretical guarantee for our methods, and present results on two real, large datasets, one on network flow data (100GB trace of 22K hosts over one month) and one on DBLP (200MB over 25 years). We show that CMD is often an order of magnitude more efficient than the state of the art (SVD and CUR): it is over 10X faster, but requires less than 1/10 of the space, for the same reconstruction accuracy. Finally, we demonstrate how CMD is used for detecting anomalies and monitoring time-evolving graphs, in which it successfully detects worm-like hierarchical scanning patterns in real network data.

Coevolving streams of numerical measurements, as well astime evolving graphs, can well be represented as tensors. Here we review the fundamental matrix and tensors tools forthe analysis and mining of large scale streams and graphs.

This paper proposes novel and effective techniques to estimate a radius to answer k-nearest neighbor queries. The first technique targets datasets where it is possible to learn the distribution about the pairwise distances between the elements, generating a global estimation that applies to the whole dataset. The second technique targets datasets where the first technique cannot be employed, generating estimations that depend on where the query center is located. The proposed k-NNF() algorithm combines both techniques, achieving remarkable speedups. Experiments performed on both real and synthetic datasets have shown that the proposed algorithm can accelerate k-NN queries more than 26 times compared with the incremental algorithm and spends half of the total time compared with the traditional k-NN() algorithms.

Given a huge online social network, how do we retrieve information from it through crawling? Even better, how do we improve the crawling performance by using parallel crawlers that work independently? In this paper, we present the framework of parallel crawlers for online social networks, utilizing a centralized queue. To show how this works in practice, we describe our implementation of the crawlers for an online auction website. The crawlers work independently, therefore the failing of one crawler does not affect the others at all. The framework ensures that no redundant crawling would occur. Using the crawlers that we built, we visited a total of approximately 11 million auction users, about 66,000 of which were completely crawled.

Given a large online network of online auction users and their histories of transactions, how can we spot anomalies and auction fraud? This paper describes the design and implementation of NetProbe, a system that we propose for solving this problem. NetProbe models auction users and transactions as a Markov Random Field tuned to detect the suspicious patterns that fraudsters create, and employs a Belief Propagation mechanism to detect likely fraudsters. Our experiments show that NetProbe is both efficient and effective for fraud detection. We report experiments on synthetic graphs with as many as 7,000 nodes and 30,000 edges, where NetProbe was able to spot fraudulent nodes with over 90% precision and recall, within a matter of seconds. We also report experiments on a real dataset crawled from eBay, with nearly 700,000 transactions between more than 66,000users, where NetProbe was highly effective at unearthing hidden networks of fraudsters, within a realistic response time of about 6 minutes. For scenarios where the underlying data is dynamic in nature, we propose IncrementalNetProbe, which is an approximate, but fast, variant of NetProbe. Our experiments prove that Incremental NetProbe executes nearly doubly fast as compared to NetProbe, while retaining over 99% of its accuracy.

How do we find a natural clustering of a real-world point set which contains an unknown number of clusters with different shapes, and which may be contaminated by noise? As most clustering algorithms were designed with certain assumptions (Gaussianity), they often require the user to give input parameters, and are sensitive to noise. In this article, we propose a robust framework for determining a natural clustering of a given dataset, based on the minimum description length (MDL) principle. The proposed framework, robust information-theoretic clustering (RIC) , is orthogonal to any known clustering algorithm: Given a preliminary clustering, RIC purifies these clusters from noise, and adjusts the clusterings such that it simultaneously determines the most natural amount and shape (subspace) of the clusters. Our RIC method can be combined with any clustering technique ranging from K-means and K-medoids to advanced methods such as spectral clustering. In fact, RIC is even able to purify and improve an initial coarse clustering, even if we start with very simple methods. In an extension, we propose a fully automatic stand-alone clustering method and efficiency improvements. RIC scales well with the dataset size. Extensive experiments on synthetic and real-world datasets validate the proposed RIC framework.

How do real graphs evolve over time? What are normal growth patterns in social, technological, and information networks? Many studies have discovered patterns in static graphs , identifying properties in a single snapshot of a large network or in a very small number of snapshots; these include heavy tails for in- and out-degree distributions, communities, small-world phenomena, and others. However, given the lack of information about network evolution over long periods, it has been hard to convert these findings into statements about trends over time. Here we study a wide range of real graphs, and we observe some surprising phenomena. First, most of these graphs densify over time with the number of edges growing superlinearly in the number of nodes. Second, the average distance between nodes often shrinks over time in contrast to the conventional wisdom that such distance parameters should increase slowly as a function of the number of nodes (like O (log n ) or O (log(log n )). Existing graph generation models do not exhibit these types of behavior even at a qualitative level. We provide a new graph generator, based on a forest fire spreading process that has a simple, intuitive justification, requires very few parameters (like the flammability of nodes), and produces graphs exhibiting the full range of properties observed both in prior work and in the present study. We also notice that the forest fire model exhibits a sharp transition between sparse graphs and graphs that are densifying. Graphs with decreasing distance between the nodes are generated around this transition point. Last, we analyze the connection between the temporal evolution of the degree distribution and densification of a graph. We find that the two are fundamentally related. We also observe that real networks exhibit this type of relation between densification and the degree distribution.

Provides instructions and guidelines to prospective authors who wish to submit manuscripts.

Multimedia and complex data are usually queried by similarity predicates. Whereas there are many works dealing with algorithms to answer basic similarity predicates, there are not generic algorithms able to efficiently handle similarity complex queries combining several basic similarity predicates. In this work we propose a simple and effective set of algorithms that can be combined to answer complex similarity queries, and a set of algebraic rules useful to rewrite similarity query expressions into an adequate format for those algorithms. Those rules and algorithms allow relational database management systems to turn complex queries into efficient query execution plans. We present experiments that highlight interesting scenarios. They show that the proposed algorithms are orders of magnitude faster than the traditional similarity algorithms. Moreover, they are linearly scalable considering the database size.

How closely related are two nodes in a graph? How to compute this score quickly, on huge, disk-resident, real graphs? Random walk with restart (RWR) provides a good relevance score between two nodes in a weighted graph, and it has been successfully used in numerous settings, like automatic captioning of images, generalizations to the "connection subgraphs", personalized PageRank, and many more. However, the straightforward implementations of RWR do not scale for large graphs, requiring either quadratic space and cubic pre-computation time, or slow response time on queries. We propose fast solutions to this problem. The heart of our approach is to exploit two important properties shared by many real graphs: (a) linear correlations and (b) block- wise, community-like structure. We exploit the linearity by using low-rank matrix approximation, and the community structure by graph partitioning, followed by the Sherman- Morrison lemma for matrix inversion. Experimental results on the Corel image and the DBLP dabasets demonstrate that our proposed methods achieve significant savings over the straightforward implementations: they can save several orders of magnitude in pre-computation and storage cost, and they achieve up to 150x speed up with 90%+ quality preservation.

Given a large social or computer network, how can we visualize it, find patterns, outliers, communities? Although several graph visualization tools exist, they cannot handle large graphs with hundred thousand nodes and possibly million edges. Such graphs bring two challenges: interactive visualization demands prohibitive processing power and, even if we could interactively update the visualization, the user would be overwhelmed by the excessive number of graphical items. To cope with this problem, we propose a formal innovation on the use of graph hierarchies that leads to GMine system. GMine promotes scalability using a hierarchy of graph partitions, promotes concomitant presentation for the graph hierarchy and for the original graph, and extends analytical possibilities with the integration of the graph partitions in an interactive environment.

How do we find a natural clustering of a real world point set, which contains an unknown number of clusters with different shapes, and which may be contaminated by noise? Most clustering algorithms were designed with certain assumptions (Gaussianity), they often require the user to give input parameters, and they are sensitive to noise. In this paper, we propose a robust framework for determining a natural clustering of a given data set, based on the minimum description length (MDL) principle. The proposed framework, Robust Information-theoretic Clustering (RIC), is orthogonal to any known clustering algorithm: given a preliminary clustering, RIC purifies these clusters from noise, and adjusts the clusterings such that it simultaneously determines the most natural amount and shape (subspace) of the clusters. Our RIC method can be combined with any clustering technique ranging from K-means and K-medoids to advanced methods such as spectral clustering. In fact, RIC is even able to purify and improve an initial coarse clustering, even if we start with very simple methods such as grid-based space partitioning. Moreover, RIC scales well with the data set size. Extensive experiments on synthetic and real world data sets validate the proposed RIC framework.

Given a huge real graph, how can we derive a representative sample? There are many known algorithms to compute interesting measures (shortest paths, centrality, betweenness, etc.), but several of them become impractical for large graphs. Thus graph sampling is essential.The natural questions to ask are (a) which sampling method to use, (b) how small can the sample size be, and (c) how to scale up the measurements of the sample (e.g., the diameter), to get estimates for the large graph. The deeper, underlying question is subtle: how do we measure success?.We answer the above questions, and test our answers by thorough experiments on several, diverse datasets, spanning thousands nodes and edges. We consider several sampling methods, propose novel methods to check the goodness of sampling, and develop a set of scaling laws that describe relations between the properties of the original and the sample.In addition to the theoretical contributions, the practical conclusions from our work are: Sampling strategies based on edge selection do not perform well; simple uniform random node selection performs surprisingly well. Overall, best performing methods are the ones based on random-walks and "forest fire"; they match very accurately both static as well as evolutionary graph patterns, with sample sizes down to about 15% of the original graph.

We present FEMine, an automatic system for image-based gene expression analysis. We perform experiments on the largest publicly available collection of Drosophila ISH (in situ hybridization) images, showing that our FEMine system achieves excellent performance in classification, clustering, and content-based image retrieval. The major innovation of FEMine is the use of automatically discovered latent spatial "themes" of gene expressions, LGEs, in the whole-embryo context, as opposed to patterns in nearly disjoint portions of an embryo proposed in previous methods.

How do we find patterns in author-keyword associations, evolving over time? Or in Data Cubes, with product-branch-customer sales information? Matrix decompositions, like principal component analysis (PCA) and variants, are invaluable tools for mining, dimensionality reduction, feature selection, rule identification in numerous settings like streaming data, text, graphs, social networks and many more. However, they have only two orders, like author and keyword, in the above example.We propose to envision such higher order data as tensors,and tap the vast literature on the topic. However, these methods do not necessarily scale up, let alone operate on semi-infinite streams. Thus, we introduce the dynamic tensor analysis (DTA) method, and its variants. DTA provides a compact summary for high-order and high-dimensional data, and it also reveals the hidden correlations. Algorithmically, we designed DTA very carefully so that it is (a) scalable, (b) space efficient (it does not need to store the past) and (c) fully automatic with no need for user defined parameters. Moreover, we propose STA, a streaming tensor analysis method, which provides a fast, streaming approximation to DTA.We implemented all our methods, and applied them in two real settings, namely, anomaly detection and multi-way latent semantic indexing. We used two real, large datasets, one on network flow data (100GB over 1 month) and one from DBLP (200MB over 25 years). Our experiments show that our methods are fast, accurate and that they find interesting patterns and outliers on the real datasets.

Given Q nodes in a social network (say, authorship network), how can we find the node/author that is the center-piece, and has direct or indirect connections to all, or most of them? For example, this node could be the common advisor, or someone who started the research area that the Q nodes belong to. Isomorphic scenarios appear in law enforcement (find the master-mind criminal, connected to all current suspects), gene regulatory networks (find the protein that participates in pathways with all or most of the given Q proteins), viral marketing and many more.Connection subgraphs is an important first step, handling the case of Q=2 query nodes. Then, the connection subgraph algorithm finds the b intermediate nodes, that provide a good connection between the two original query nodes.Here we generalize the challenge in multiple dimensions: First, we allow more than two query nodes. Second, we allow a whole family of queries, ranging from 'OR' to 'AND', with 'softAND' in-between. Finally, we design and compare a fast approximation, and study the quality/speed trade-off.We also present experiments on the DBLP dataset. The experiments confirm that our proposed method naturally deals with multi-source queries and that the resulting subgraphs agree with our intuition. Wall-clock timing results on the DBLP dataset show that our proposed approximation achieve good accuracy for about 6:1 speedup.

Efficiently and accurately searching for similarities among time series and discovering interesting patterns is an important and non-trivial problem. In this paper, we introduce a new representation of time series, the multiresolution vector quantized (MVQ) approximation, along with a new distance function. The novelty of MVQ is that it keeps both local and global information about the original time series in a hierarchical mechanism, processing the original time series at multiple resolutions. Moreover, the proposed representation is symbolic employing key subsequences and potentially allows the application of text-based retrieval techniques into the similarity analysis of time series. The proposed method is fast and scales linearly with the size of database and the dimensionality. Contrary to the vast majority in the literature that uses the Euclidean distance, MVQ uses a multi-resolution/hierarchical distance function. We performed experiments with real and synthetic data. The proposed distance function consistently outperforms all the major competitors (Euclidean, dynamic time warping, piecewise aggregate approximation) achieving up to 20% better precision/recall and clustering accuracy on the tested datasets.

We have introduced the problem of automatic lag correlation detection on streaming data and proposed AutoLag to address this problem by using careful approximations and smoothing. Our experiments on real and realistic data show that AutoLag works as expected, estimating the unknown lags with excellent accuracy and significant speed-up. In our experiments on real and realistic data, AutoLag was up to about 42,000 times faster than the naive implementation, with at most 1% relative error.