Vision models fail both under tiny pixel attacks and under real-world shifts in style or background because they latch onto spurious features. We propose a two-step, label-free method. (1) Spurious-Invariant Self-Supervised Pre-training (SISSP) trains an encoder to collapse representations of the same object despite randomized styles and backgrounds, pruning shortcut signals. (2) Semantic-Alignment Adversarial Refinement (SAAR) takes any attack and projects it back into a small ball within SISSP feature space, yielding adversaries that look natural yet still fool the classifier. Fine-tuning with SISSP features and SAAR images produces a ResNet-50 that retains 64% ImageNet accuracy, 46% PGD robustness without environment labels or specialized augmentations. Together, SISSP provides a semantics-aware metric and SAAR generates on-manifold adversaries, achieving the first ImageNet-scale model robust to both pixel-level noise and semantic shifts.

Temporal Graph Neural Networks (TGNNs) have emerged as powerful tools for modeling dynamic interactions across various domains. The design space of TGNNs is notably complex, given the unique challenges in runtime efficiency and scalability raised by the evolving nature of temporal graphs. We contend that many of the existing works on TGNN modeling inadequately explore the design space, leading to suboptimal designs. Viewing TGNN models through a performance-focused lens often obstructs a deeper understanding of the advantages and disadvantages of each technique. Specifically, benchmarking efforts inherently evaluate models in their original designs and implementations, resulting in unclear accuracy comparisons and misleading runtime. To address these shortcomings, we propose a practical comparative evaluation framework that performs a design space search across well-known TGNN modules based on a unified, optimized code implementation. Using our framework, we make the first efforts towards addressing three critical questions in TGNN design, spending over 10,000 GPU hours: (1) investigating the efficiency of TGNN module designs, (2) analyzing how the effectiveness of these modules correlates with dataset patterns, and (3) exploring the interplay between multiple modules. Key outcomes of this directed investigative approach include demonstrating that the most recent neighbor sampling and attention aggregator outperform uniform neighbor sampling and MLP-Mixer aggregator; Assessing static node memory as an effective node memory alternative, and showing that the choice between static or dynamic node memory should be based on the repetition patterns in the dataset. Our in-depth analysis of the interplay between TGNN modules and dataset patterns should provide a deeper insight into TGNN performance along with potential research directions for designing more general and effective TGNNs.

Graph neural networks (GNNs) have gained significant attention across diverse areas due to their superior performance in learning graph representations. While GNNs exhibit superior performance compared to other methods, they are primarily designed for homogeneous graphs, where all nodes and edges are of the same type. Training a GNN model for large-scale graphs incurs high computation and storage costs, especially when considering the heterogeneous structural information of each node. To address the demand for efficient GNN training, various sampling methods have been proposed. In this paper, we propose a sampling method based on bandit sampling, an online learning algorithm with provable convergence under weak assumptions on the learning objective. To the best of our knowledge, this is the first bandit-based sampling method applied to heterogeneous GNNs with a theoretical guarantee. The main idea is to prioritize node types with more informative connections with respect to the learning objective. Compared with existing techniques for GNN training on heterogeneous graphs, extensive experiments using the Open Academic Graph (OAG) dataset demonstrate that our proposed method outperforms the state-of-the-art in terms of the runtime across various tasks with a speed-up of 1.5-2x, while achieving similar accuracy.

Multi-Agent Reinforcement Learning (MARL) has achieved significant success in large-scale AI systems and big-data applications such as smart grids, surveillance, etc. Existing advancements in MARL algorithms focus on improving the rewards obtained by introducing various mechanisms for inter-agent cooperation. However, these optimizations are usually compute- and memory-intensive, thus leading to suboptimal speed performance in end-to-end training time. In this work, we analyze the speed performance (i.e., latency-bounded throughput) as the key metric in MARL implementations. Specifically, we first introduce a taxonomy of MARL algorithms from an acceleration perspective categorized by (1) training scheme and (2) communication method. Using our taxonomy, we identify three state-of-the-art MARL algorithms - Multi-Agent Deep Deterministic Policy Gradient (MADDPG), Target-oriented Multi-agent Communication and Cooperation (ToM2C), and Networked Multi-Agent RL (NeurComm) - as target benchmark algorithms, and provide a systematic analysis of their performance bottlenecks on a homogeneous multi-core CPU platform. We justify the need for MARL latency-bounded throughput to be a key performance metric in future literature while also addressing opportunities for parallelization and acceleration.

Temporal Knowledge Graphs store events in the form of subjects, relations, objects, and timestamps which are often represented by dynamic heterogeneous graphs. Event forecasting is a critical and challenging task in Temporal Knowledge Graph reasoning that predicts the subject or object of an event in the future. To obtain temporal embeddings multi-step away in the future, existing methods learn generative models that capture the joint distribution of the observed events. To reduce the high computation costs, these methods rely on unrealistic assumptions of independence and approximations in training and inference. In this work, we propose SeDyT, a discriminative framework that performs sequence modeling on the dynamic entity embeddings to solve the multi-step event forecasting problem. SeDyT consists of two components: a Temporal Graph Neural Network that generates dynamic entity embeddings in the past and a sequence model that predicts the entity embeddings in the future. Compared with the generative models, SeDyT does not rely on any heuristic-based probability model and has low computation complexity in both training and inference. SeDyT is compatible with most Temporal Graph Neural Networks and sequence models. We also design an efficient training method that trains the two components in one gradient descent propagation. We evaluate the performance of SeDyT on five popular datasets. By combining temporal Graph Neural Network models and sequence models, SeDyT achieves an average of 2.4% MRR improvement when not using the validation set and more than 10% MRR improvement when using the validation set.

Graph Neural Networks (GNNs) are proven to be powerful models to generate node embedding for downstream applications. However, due to the high computation complexity of GNN inference, it is hard to deploy GNNs for large-scale or real-time applications. In this paper, we propose to accelerate GNN inference by pruning the dimensions in each layer with negligible accuracy loss. Our pruning framework uses a novel LASSO regression formulation for GNNs to identify feature dimensions (channels) that have high influence on the output activation. We identify two inference scenarios and design pruning schemes based on their computation and memory usage for each. To further reduce the inference complexity, we effectively store and reuse hidden features of visited nodes, which significantly reduces the number of supporting nodes needed to compute the target embedding. We evaluate the proposed method with the node classification problem on five popular datasets and a real-time spam detection application. We demonstrate that the pruned GNN models greatly reduce computation and memory usage with little accuracy loss. For full inference, the proposed method achieves an average of 3.27X speedup with only 0.002 drop in F1-Micro on GPU. For batched inference, the proposed method achieves an average of 6.67X speedup with only 0.003 drop in F1-Micro on CPU. To the best of our knowledge, we are the first to accelerate large scale real-time GNN inference through channel pruning.

Tip decomposition is a crucial kernel for mining dense subgraphs in bipartite networks, with applications in spam detection, analysis of affiliation networks etc. It creates a hierarchy of vertex-induced subgraphs with varying densities determined by the participation of vertices in butterflies (2, 2-bicliques). To build the hierarchy, existing algorithms iteratively follow a delete-update (peeling) process: deleting vertices with the minimum number of butterflies and correspondingly updating the butterfly count of their 2-hop neighbors. The need to explore 2-hop neighborhood renders tip-decomposition computationally very expensive. Furthermore, the inherent sequentiality in peeling only minimum butterfly vertices makes derived parallel algorithms prone to heavy synchronization. In this paper, we propose a novel parallel tip-decomposition algorithm - REfine CoarsE-grained Independent Tasks (RECEIPT) that relaxes the peeling order restrictions by partitioning the vertices into multiple independent subsets that can be concurrently peeled. This enables RECEIPT to simultaneously achieve a high degree of parallelism and dramatic reduction in synchronizations. Further, RECEIPT employs a hybrid peeling strategy along with other optimizations that drastically reduce the amount of wedge exploration and execution time. We perform detailed experimental evaluation of RECEIPT on a shared-memory multicore server. It can process some of the largest publicly available bipartite datasets orders of magnitude faster than the state-of-the-art algorithms - achieving up to 1100× and 64× reduction in the number of thread synchronizations and traversed wedges, respectively. Using 36 threads, RECEIPT can provide up to 17.1× self-relative speedup.

To improve the reliability of content shared on social media, effective strategies for mitigating the diffusion of fake news are increasingly necessary. Traditionally, to counter false belief a competing cascade approach is used. This approach assumes that the opposite belief is already known, and thus, not applicable to newly spreading fake news. Another approach is to block nodes and links of the network to impede the flow of fake news (rumor/influence blocking). However, a more active way to battle the dissemination of fake news is to propagate the corresponding real news, since people who receive the real news in tandem with the fake news are less likely to believe in fake news. Such a setting is especially useful on a messaging platform such as WhatsApp, where the news item flows as a private message and the correction of fake news and its propagation must be performed by the users within the network as they receive it. To achieve this goal, we propose network-based recruitment for active fake news correction (RecANt) to find a set of individuals of a pre-defined size to be incentivized for actively fact-checking and passing on the real news so as to reach the maximum number of nodes in the network. These individuals should be such that they are likely to receive the fake news so that they can test its credibility, and when they propagate the corresponding real news, it reaches a large number of individuals. We prove that RecANt is NP-Hard with a monotone and submodular objective, leading to a polynomial time greedy algorithm (AFC) which provides a (1 - 1/e - ε)-approximation. We further optimize the runtime of AFC by developing a fast graph-pruning heuristic (RAFC) that performs as well as AFC in checking the spread of fake news while reducing the runtime significantly. Simulations on several networks demonstrate that our approach outperforms popular social network centrality measures and state-of-the-art information diffusion algorithm.

Accurate prediction of crime incidents can assist the police in better planning of prevention strategies and scheduling deployment. The problem is often studied as a spatio-temporal regression problem approached by dividing the area of interest into a grid of uniform cells, and performing regression on timeseries of each cell. We propose that changing the method of division of the area can significantly improve crime prediction. We demonstrate this using a heterogeneous division of the area obtained by our partitioning algorithm that takes into account the density of crime. We further show that existing measures do not provide a fair comparison of two methods that partition the area in two different ways. To address this severe drawback in crime prediction evaluation, we propose a novel measure which is based on optimal allocation of resources relying on the prediction and then checking the actual number of crimes that would have been avoided by the allocation. Essentially, our measure answers the question of which model would have assisted in preventing most number of actual crimes if allocation were to be done using the predicted crimes. We also prove that a greedy algorithm results in the optimal allocation resources, thus making our evaluation computationally lightweight. Experiments on real-world datasets demonstrate that heterogeneous division of the area results in improved crime prediction while drastically decreasing the number of models to be trained compared to uniform grid division.

Hub labeling is widely used to improve the latency and throughput of Point-to-Point Shortest Distance (PPSD) queries in graph databases. However, constructing hub labeling, even via the state-of-the-art Pruned Landmark Labeling (PLL) algorithm is computationally intensive. PLL further has a sequential root order label dependency that makes it challenging to parallelize. Hence, the existing parallel approaches are often plagued by label size increase, poor scalability and inability to process large weighted graphs. In this paper, we develop novel algorithms that construct the minimal (guaranteed) Canonical Hub Labeling on shared and distributed-memory parallel systems in a scalable and efficient manner. Our key contribution, the PLaNT algorithm, provides an embarrassingly parallel approach for label construction that scales well beyond the limits of current practice. Our approach is the first to employ a collaborative label partitioning scheme across multiple nodes of a cluster, for completely in-memory labeling and parallel querying on massive graphs whose labels cannot fit on a single node. On a single node with 72-threads, our shared-memory algorithm is up to 47.4X faster than sequential PLL. While our labeling time is comparable to the state-of-the-art shared-memory paraPLL, our label size is 17% smaller on average. PLaNT demonstrates superior parallel scalability. It can process significantly larger graphs and construct labeling orders of magnitude faster than the state-of-the-art distributed paraPLL. Compared to the best shared-memory parallel algorithm, it achieves up to 9.5X speedup on a 64 node cluster.

In developing countries, reliable electricity access is often undermined by the absence of supply from the national power grid and/or load shedding. To alleviate this problem, smart micro-grid (SMG) networks that are small scale distributed electricity provision networks composed of individual electricity providers and consumers, are being increasingly deployed. To ensure the reliable operation of SMGs, monitoring is necessary for data collection and state estimation processes. However, highly calibrated and trustworthy smart meters that are ideally suited to perform such monitoring tasks are often costly and non-ideally suited to SMGs which operate under unreliable communication network infrastructures. As a result, SMGs are an easy target to an adversary who can very easily gain access to private information by monitoring transmission between nodes in the SMG network, and launch inference-based privacy attacks. These attacks lead to electricity theft and grid instability problems in the SMG. The widely popular differential privacy (DP) technique (a rigorous technique in the family of privacy-preserving data publishing (PPDP) techniques to mathematically guarantee the preservation of data privacy) does not address multi-attribute correlations, that are inherently exploited by an adversary in inference attacks. In this paper, we propose HIDE, an oblivious computationally efficient, and rigorous information-theoretic privacy engineering framework for datasets/databases arising in the SMG environments that robustly accounts for multi-attribute correlations while preserving data privacy in a provably optimal fashion. A salient and powerful advantage of HIDE is its ability to generate optimal utility-privacy tradeoffs (computationally efficiently) when the privacy preserving entity in the worst case might have no prior statistical information that links a user's private data with his public data.

Interventions to reduce violence among homeless youth are difficult to implement due to the complex nature of violence. However, a peer-based intervention approach would likely be a worthy approach as it has been shown that individuals who interact with more violent individuals are more likely to be violent, suggesting a contagious nature of violence. We propose Uncertain Voter Model to represent the complex process of diffusion of violence over a social network, that captures uncertainties in links and time over which the diffusion of violence takes place. Assuming this model, we define Violence Minimization problem where the task is to select a predefined number of individuals for intervention so that the expected number of violent individuals in the network is minimized over a given time-frame. We extend the problem to a probabilistic setting, where the success probability of converting an individual into non-violent is a function of the number of “units” of intervention performed on them. We provide algorithms for finding the optimal intervention strategies for both scenarios. We demonstrate that our algorithms perform significantly better than interventions based on popular centrality measures in terms of reducing violence.

In recent years, deep convolutional networks have been successfully used for the tasks of image classification and speech recognition. The highly non-linear modeling combined with its emphasis on local connectivity makes them highly suitable for such tasks. However, their performance in other domains is not well explored. Specifically, in the oil industry, researchers use manual features from time series data as input to various machine learning models. In this paper, we employ deep convolutional autoencoders to extract non linear latent features from time series data. We propose a novel deep network architecture and show its efficacy in two oil field tasks related to reservoir optimization - steam job prediction and slippage detection. We show that our architecture outperforms state-of-the-art methods significantly on steam job prediction. We demonstrate the success of our model on an oil field dataset which consists of production and failure data of over two years. Our architecture achieves a precision of 98% for precision@50 in steam job prediction, and 25% improvement over the methods used in the industry. To the best of our knowledge, we are the first to attempt to automatically detect slippage failures in well pumps. We are able to classify slippage events with 70.3% accuracy, a 10.6% improvement over using manually defined input features.

Subgraph isomorphism is a fundamental graph problem with many applications. Due to its NP-Hard nature, subgraph isomorphism in large dynamic graphs is considered as a challenging problem. In this paper, we present a distributed graph pruning algorithm (D-IDS) for dynamic graphs to enable efficient subgraph isomorphism. D-IDS continuously maintains the maximum dual simulation match in a dynamic graph. We develop D-ISI, a distributed incremental algorithm for subgraph isomorphism that utilizes D-IDS. We evaluated our algorithms on a commodity cluster in Amazon EC2 using real world graph datasets. Our evaluation results show that the graph pruning technique is highly effective on graphs with small diameter where it achieves over 60% reduction in graph size.

Often in marketing, political campaigns and social media, two competing products or opinions propagate over a social network. Studying social influence in such competing cascades scenarios enables building effective strategies for maximizing the propagation of one process by targeting the most "influential" nodes in the network. The majority of prior work however, focuses on unsigned networks where individuals adopt the opinion of their neighbors with certain probability. In real life, relationships between individuals can be positive (e.g., friend of relationship) or negative (e.g. connection between "foes"). According to social theory, people tend to have similar opinions to their friends but opposite of their foes. In this work, we study the problem of competing cascades on signed networks, which has been relatively unexplored. Particularly, we study the progressive propagation of two competing cascades in a signed network under the Independent Cascade Model, and provide an approximate analytical solution to compute the probability of infection of a node at any given time. We leverage our analytical solution to the problem of competing cascades in signed networks to develop a heuristic for the influence maximization problem. Unlike prior work, we allow the seed-set to be initialized with populations of both cascades with the end goal of maximizing the spread of one cascade. We validate our approach on several large-scale real-world and synthetic networks. Our experiments demonstrate that our influence maximization heuristic significantly outperforms state-of-the-art methods, particularly when the network is dominated by distrust relationships.

The pervasive deployment of advanced sensing infrastructure in Cyber-Physical systems, such as the Smart Grid, has resulted in an unprecedented data explosion. Such data exhibit both large volumes and high velocity characteristics, two of the three pillars of Big Data, and have a time-series notion as datasets in this context typically consist of successive measurements made over a time interval. Time-series data can be valuable for data mining and analytics tasks such as identifying the "right" customers among a diverse population, to target for Demand Response programs. However, time series are challenging to mine due to their high dimensionality. In this paper, we motivate this problem using a real application from the smart grid domain. We explore novel representations of time-series data for BigData analytics, and propose a clustering technique for determining natural segmentation of customers and identification of temporal consumption patterns. Our method is generizable to large-scale, real-world scenarios, without making any assumptions about the data. We evaluate our technique using real datasets from smart meters, totaling ∼ 18,200,000 data points, and show the efficacy of our technique in efficiency detecting the number of optimal number of clusters.

The performance of prediction models is often based on “abstract metrics” that estimate the model's ability to limit residual errors between the observed and predicted values. However, meaningful evaluation and selection of prediction models for end-user domains requires holistic and application-sensitive performance measures. Inspired by energy consumption prediction models used in the emerging “big data” domain of Smart Power Grids, we propose a suite of performance measures to rationally compare models along the dimensions of scale independence, reliability, volatility and cost. We include both application independent and dependent measures, the latter parameterized to allow customization by domain experts to fit their scenario. While our measures are generalizable to other domains, we offer an empirical analysis using real energy use data for three Smart Grid applications: planning, customer education and demand response, which are relevant for energy sustainability. Our results underscore the value of the proposed measures to offer a deeper insight into models' behavior and their impact on real applications, which benefit both data mining researchers and practitioners.

Understanding how information flows in online social networks is of great importance. It is generally difficult to obtain accurate prediction results of cascades over such networks, therefore a variety of diffusion models have been proposed in the literature to simulate diffusion processes instead. We argue that such models require extensive simulation results to produce good estimates of future spreads. In this work, we take a complimentary approach. We present a generalized, analytical model of influence in social networks that captures social influence at various levels of granularity, ranging from pairwise influence, to local neighborhood, to the general population, and external events, therefore capturing the complex dynamics of human behavior. We demonstrate that our model can integrate a variety of diffusion models. Particularly, we show that commonly used diffusion models in social networks can be reduced to special cases of our model, by carefully defining their parameters. Our goal is to provide a closed-form expression to approximate the probability of infection for every node in an arbitrary, directed network at any time t. We quantitatively evaluate the approximation quality of our analytical solution as compared to numerous popular diffusion models on a real-world dataset and a series of synthetic graphs.

Big data applications such as in smart electric grids, transportation, and remote environment monitoring involve geographically dispersed sensors that periodically send back information to central nodes. In many cases, data from sensors is not available at central nodes at a frequency that is required for real-time modeling and decision-making. This may be due to physical limitations of the transmission networks, or due to consumers limiting frequent transmission of data from sensors located at their premises for security and privacy concerns. Such scenarios lead to partial data problem and raise the issue of data veracity in big data applications. We describe a novel solution to the problem of making short term predictions (up to a few hours ahead) in absence of real-time data from sensors in Smart Grid. A key implication of our work is that by using real-time data from only a small subset of influential sensors, we are able to make predictions for all sensors. We thus reduce the communication complexity involved in transmitting sensory data in Smart Grids. We use real-world electricity consumption data from smart meters to empirically demonstrate the usefulness of our method. Our dataset consists of data collected at 15-min intervals from 170 smart meters in the USC Microgrid for 7 years, totaling 41,697,600 data points.

Smart grids are becoming popular with the advent of sophisticated smart meters. They allow utilities to optimize energy consumption during peak hours by applying various demand response techniques including voluntary curtailment, direct control and price incentives. To sustain the curtailment over long periods of time of up to several hours utilities need to make fast and accurate consumption predictions on a large set of customers based on a continuous flow of real time data and huge historical data sets. Given the numerous consumption patterns customers exhibit, different prediction methods need to be used to reduce the prediction error. The straightforward approach of testing each customer against every method is unfeasible in this large volume and high velocity environment. To this aim, we propose a neural network based approach for automatically selecting the best prediction method per customer by relying only on a small subset of customers. We also introduce two historical averaging methods for consumption prediction that take advantage of the variability of the data and continuously update the results based on a sliding window technique. We show that once trained, the proposed neural network does not require frequent retraining, ensuring its applicability in online scenarios such as the sustainable demand response.

We study the problem of identifying discriminative features in Big Data arising from heterogeneous sensors. We highlight the heterogeneity in sensor data from engineering applications and the challenges involved in automatically extracting only the most interesting features from large datasets. We formulate this problem as that of classification of multivariate time series and design shapelet-based algorithms for this task. We design a novel approach, called Shapelet Forests (SF), which combines shapelet extraction with feature selection. We evaluate our proposed method with other approaches for mining shapelets from multivariate time series using data from real-world engineering applications. Quantitative analysis of the experiments shows that SF performs better than the baseline approaches and achieves high classification accuracy. In addition, the method enables identification of noisy sensors from multivariate data and discounts their use for classification.

Popular social networking sites have revolutionized the way people interact on the Web, enabling rapid information dissemination and search. In an enterprise, understanding how information flows within and between organizational levels and business units is of great importance. Despite numerous studies in information diffusion in online social networks, little is known about factors that affect the dynamics of technological adoption at the workplace. Here, we address this problem, by examining the impact of organizational hierarchy in adopting new technologies in the enterprise. Our study suggests that middle-level managers are more successful in influencing employees into adopting a new microblogging service. Further, we reveal two distinct patterns of peer pressure, based on which employees are not only more likely to adopt the service, but the rate at which they do so quickens as the popularity of the new technology increases. We integrate our findings into two intuitive, realistic agent-based computational models that capture the dynamics of adoption at both microscopic and macroscopic levels. We evaluate our models in a real-world dataset we collected from a multinational Fortune 500 company. Prediction results show that our models provide great improvements over commonly used diffusion models. Our findings provide significant insights to managers seeking to realize the dynamics of adoption of new technologies in their company, and could assist in designing better strategies for rapid and efficient technology adoption and information dissemination at the workplace.

To enhance human resource management and personalized information acquisition, employee ontology is used to model business concepts and relations between them for enterprises. In this paper, we propose an employee ontology that integrates user static properties from formal structures with dynamic interests and expertise extracted from informal communication signals. We mine user's interests at both personal and professional level from informal interactions on communication platforms at the workplace. We show how complex semantic queries enable granular analysis. At the microscopic level, enterprises can utilize the results to better understand how their employees work together to complete tasks or produce innovative ideas, identify experts and influential individuals. At the macroscopic level, conclusions can be drawn, among others, about collective behavior and expertise in varying granularities (i.e. single employee to the company as a whole).

Emerging Big Data applications in areas like ecommerce and energy industry require both online and on-demand queries to be performed over vast and fast data arriving as streams. These present novel challenges to Big Data management systems. Complex Event Processing (CEP) is recognized as a high performance online query scheme which in particular deals with the velocity aspect of the 3-V's of Big Data. However, traditional CEP systems do not consider data variety and lack the capability to embed ad hoc queries over the volume of data streams. In this paper, we propose H2O, a stateful complex event processing framework, to support hybrid online and on-demand queries over realtime data. We propose a semantically enriched event and query model to address data variety. A formal query algebra is developed to precisely capture the stateful and containment semantics of online and on-demand queries. We describe techniques to achieve the interactive query processing over realtime data featured by efficient online querying, dynamic stream data persistence and on-demand access. The system architecture is presented and the current implementation status reported.

Social networks have become a popular medium for people to communicate and distribute ideas, content, news, and advertisements. Social content annotation has naturally emerged as a method of categorization and filtering of online information. The unrestricted vocabulary users choose from to annotate content has often lead to an explosion of the size of space in which search is performed. In this article, we propose latent topic models as a principled way of reducing the dimensionality of such data and capturing the dynamics of collaborative annotation process. We propose three generative processes to model latent user tastes with respect to resources they annotate with metadata. We show that latent user interests combined with social clues from the immediate neighborhood of users can significantly improve social link prediction in the online music social media site Last.fm. Most link prediction methods suffer from the high class imbalance problem, resulting in low precision and/or recall. In contrast, our proposed classification schemes for social link recommendation achieve high precision and recall with respect to not only the dominant class (nonexistence of a link), but also with respect to sparse positive instances, which are the most vital in social tie prediction.

Popular social networking sites have revolutionized the way people interact on the Web. Researchers have studied social networks from numerous perspectives, mostly focusing on publicly available social networks and microblogging sites. Enterprises however have recently being adopting and utilizing microblogging services as part of their day to day operations. The goal of this paper is to study the topological properties of a corporate microblogging service, its dynamics and characteristics. Through an extensive analysis of enterprise microblogging data, we provide insights on the structural properties of the extracted network of directed messages sent between users of a corporate microblogging service, as well as the lexical and topical alignment of users. We compare our results to traditional, general purpose, online social networks and discuss the implications of our findings. To the best of our knowledge, this work is the first quantitative study of an enterprise microblogging service, its usage characteristics, and its derived social network based on replies between users.

Data availability in online social networks as well as the business world has lately not been an issue. Vast amounts of data are being generated by social networking users in the form of informal interactions. What has been an issue, is the transformation of data into useful information, that in time and with appropriate processing becomes knowledge. In this paper we examine knowledge generation under informal social communications, based on semantically enriched user-generated data and associated metadata. We dynamically capture users' interests and expertise using such semantically enriched content. Knowledge networks of users emerge, exhibiting collective intelligence. To capture such collective knowledge, we propose a novel knowledge base paradigm, which seamlessly integrates information from multiple platforms and facilitates knowledge extraction, mining, discovery and inferencing. Using semantically enriched user profiles, we compute semantic similarity between users and content in a joint semantic space, driving numerous applications.

This paper presents an automatic method for understanding and interpreting the semantics of unannotated web images. We observe that the relations between objects in an image carry important semantics about the image. To capture and describe such semantics, we propose Object Relation Network (ORN), a graph model representing the most probable meaning of the objects and their relations in an image. Guided and constrained by an ontology, ORN transfers the rich semantics in the ontology to image objects and the relations between them, while maintaining semantic consistency (e.g., a soccer player can kick a soccer ball, but cannot ride it). We present an automatic system which takes a raw image as input and creates an ORN based on image visual appearance and the guide ontology. We demonstrate various useful web applications enabled by ORNs, such as automatic image tagging, automatic image description generation, and image search by image.

This paper presents a novel ranking method for complex semantic relationship (semantic association) search based on user preferences. The authors’ method employs a learning-to-rank algorithm to capture each user’s preferences. Using this, it automatically constructs a personalized ranking function for the user. The ranking function is then used to sort the results of each subsequent query by the user. Query results that more closely match the user’s preferences gain higher ranks. Their method is evaluated using a real-world RDF knowledge base created from Freebase linked-open-data. The experimental results show that the authors’ method significantly improves the ranking quality in terms of capturing user preferences, compared with the state-of-the-art.

Data management and analysis has become an integral component in the area of reservoir engineering. An important metric that determines the overall effectiveness of data analysis is data quality. Data provenance, the metadata that pertains to the derivation history of data objects, has emerged as an invaluable asset in evaluating data quality. The reservoir facilities and software systems that collect provenance information are often distributed, thus making it difficult to analyze provenance data. Our primary contribution in this paper is an approach for provenance information integration in reservoir engineering.