Text-attributed graphs (TAGs), which combine structural and textual node information, are ubiquitous across many domains. Recent work integrates Large Language Models (LLMs) with Graph Neural Networks (GNNs) to jointly model semantics and structure, resulting in more general and expressive models that achieve state-of-the-art performance on TAG benchmarks. However, this integration introduces dual vulnerabilities: GNNs are sensitive to structural perturbations, while LLM-derived features are vulnerable to prompt injection and adversarial phrasing. While existing adversarial attacks largely perturb structure or text independently, we find that uni-modal attacks cause only modest degradation in LLM-enhanced GNNs. Moreover, many existing attacks assume unrealistic capabilities, such as white-box access or direct modification of graph data. To address these gaps, we propose GraphTextack, the first black-box, multi-modal node injection attack for LLM-enhanced GNNs. GraphTextack injects nodes with carefully crafted structure and semantics to degrade model performance, operating under a realistic threat model without relying on model internals or surrogate models. To navigate the combinatorial, non-differentiable search space of connectivity and feature assignments, GraphTextack introduces a novel evolutionary optimization framework with a multi-objective fitness function that balances local prediction disruption and global graph influence. Extensive experiments on five datasets and two state-of-the-art LLM-enhanced GNN models show that GraphTextack significantly outperforms 12 strong baselines.

Generative retrieval (GR) has revolutionized recommendation systems by integrating the content of the items into semantic identifiers. However, existing frameworks predominantly isolate modalities (e.g., relying solely on text), overlooking the inherently multimodal nature of real-world items. This work addresses the underexplored challenge of Multimodal Generative Retrieval (MGR). Through a systematic analysis of Early and Late Fusion strategies, we reveal that naive integration fails due to two critical limitations: modality sensitivity, where one modality dominates the representation, and modality correspondence, where the model fails to align distinct semantic IDs across modalities. To overcome these challenges, we introduce MGR-LF++, an enhanced late fusion framework. MGR-LF++ employs contrastive modality alignment to enforce cross-modal consistency and incorporates special tokens to preserve autoregressive integrity. Extensive experiments demonstrate that MGR-LF++ achieves performance improvements of over 20% compared to unimodal and naive multimodal alternatives.

Recent agent-based recommendation frameworks aim to simulate user behaviors by incorporating memory mechanisms and prompting strategies, but they struggle with hallucinating non-existent items and full-catalog ranking. Besides, a largely underexplored opportunity lies in leveraging LLMs' commonsense reasoning to capture user intent through substitute and complement relationships between items, which are usually implicit in datasets and difficult for traditional ID-based recommenders to capture. In this work, we propose a novel LLM-agent framework, AgentDR, which bridges LLM reasoning with scalable recommendation tools. Our approach delegates full-ranking tasks to traditional models while utilizing LLMs to (i) integrate multiple recommendation outputs based on personalized tool suitability and (ii) reason over substitute and complement relationships grounded in user history. This design mitigates hallucination, scales to large catalogs, and enhances recommendation relevance through relational reasoning. Through extensive experiments on three public grocery datasets, we show that our framework achieves superior full-ranking performance, yielding on average a twofold improvement over its underlying tools. We also introduce a new LLM-based evaluation metric that jointly measures semantic alignment and ranking correctness.

Homophily, as a measure, has been critical to increasing our understanding of graph neural networks (GNNs). However, to date this measure has only been analyzed in the context of static graphs. In our work, we explore homophily in dynamic settings. Focusing on graph convolutional networks (GCNs), we demonstrate theoretically that in dynamic settings, current GCN discriminative performance is characterized by the probability that a node’s future label is the same as its neighbors’ current labels. Based on this insight, we propose dynamic homophily, a new measure of homophily that applies in the dynamic setting. This new measure correlates with GNN discriminative performance and sheds light on how to potentially design more powerful GNNs for dynamic graphs. Leveraging a variety of dynamic node classification datasets, we demonstrate that popular GNNs are not robust to low dynamic homophily. Going forward, our work represents an important step towards understanding homophily and GNN performance in dynamic node classification.

In this work, we theoretically demonstrate that current graph positional encodings (PEs) are not beneficial and could potentially hurt performance in tasks involving heterophilous graphs, where nodes that are close tend to have different labels. This limitation is critical as many real-world networks exhibit heterophily, and even highly homophilous graphs can contain local regions of strong heterophily. To address this limitation, we propose Learnable Laplacian Positional Encodings (LLPE), a new PE that leverages the full spectrum of the graph Laplacian, enabling them to capture graph structure on both homophilous and heterophilous graphs. Theoretically, we prove LLPE’s ability to approximate a general class of graph distances and demonstrate its generalization properties. Empirically, our evaluation on 12 benchmarks demonstrates that LLPE improves accuracy across a variety of GNNs, including graph transformers, by up to 35% and 14% on synthetic and real-world graphs, respectively. Going forward, our work represents a significant step towards developing PEs that effectively capture complex structures in heterophilous graphs.

Inspired by the success of Large Language Models (LLMs) in language and vision tasks, there has been growing interest in applying LLMs to graph tasks, particularly on Text-Attributed Graphs (TAGs). However, most prior work tackles the node classification task. In this work, we evaluate an LLM's ability to reason over structured data and infer new facts based on learned patterns by focusing on link prediction (LP)-the task of predicting missing links between nodes-that is understudied in the literature. This task poses two key challenges: (1) How to effectively integrate pairwise structural information, which is crucial for LP performance, into LLMs, and (2) how to address the computational bottleneck during inference. To tackle these challenges, we propose LinkGPT, the first LLM-based training and inference framework specifically designed for LP on homogeneous TAGs. To enhance the LLM's ability to understand the underlying structure, we carefully design a node encoder and pairwise encoder, and leverage a two-stage instruction tuning to effectively incorporate the nodewise and pairwise information into LLMs. For inference efficiency, we introduce a retrieval-reranking scheme. Extensive experiments show that LinkGPT achieves state-of-the-art performance on real-world graphs and demonstrates superior zero-shot and few-shot generalization. At inference time, it achieves a 10× speedup while maintaining high LP accuracy.

Graph machine learning has made significant strides in recent years, yet the integration of visual information with graph structure and its potential for improving performance in downstream tasks remains an underexplored area. To address this critical gap, we introduce the Multimodal Graph Benchmark (MM-Graph), a pioneering benchmark that incorporates both visual and textual information into graph learning tasks. MM-Graph extends beyond existing text-attributed graph benchmarks, offering a more comprehensive evaluation framework for multimodal graph learning. Our benchmark comprises seven diverse datasets of varying scales (ranging from thousands to millions of edges), designed to assess algorithms across different tasks in real-world scenarios. These datasets feature rich multimodal node attributes, including visual data, which enables a more holistic evaluation of various graph learning frameworks in complex, multimodal environments. To support advancements in this emerging field, we provide an extensive empirical study on various graph learning frameworks when presented with features from multiple modalities, particularly emphasizing the impact of visual information. This study offers valuable insights into the challenges and opportunities of integrating visual data into graph learning.

Large Generative Models (LGMs) such as GPT, Stable Diffusion, Sora, and Suno are trained on a huge amount of texts, images, videos, and audio that are extremely diverse from numerous domains. This large-scale training paradigm on diverse well-curated data enhances the creativity and diversity of the generated content. However, all previous graph-generative models (e.g., GraphRNN, MDVAE, MoFlow, GDSS, and DiGress) have been trained only on one dataset each time, which cannot replicate the revolutionary success achieved by LGMs in other fields. To remedy this crucial gap, we propose a large-scale training paradigm that uses a large corpus of graphs (over 5000 graphs) from 13 domains, leading to the development of large graph generative models (LGGMs). We empirically demonstrate that the pre-trained LGGMs have superior zero-shot generative capability to existing graph generative models. Furthermore, our pre-trained LGGMs can be easily fine-tuned with graphs from target domains and demonstrate even better performance than those directly trained from scratch, behaving as a solid starting point for real-world customization. Inspired by Stable Diffusion, we further equip LGGMs with the Text-to-Graph generation capability, such as providing the description of the network name and domain (i.e., "The power-1138-bus graph represents a network of buses in a power distribution system.") and network statistics (i.e., "The graph has a low average degree, suitable for modeling social media interactions."). This Text-to-Graph capability integrates the extensive world knowledge in the underlying language model, offering users fine-grained control of the generated graphs. We release the code, the model checkpoint, and the datasets at https://github.com/KINDLab-Fly/LGGM.

Frontier large language models (LLMs) have demonstrated remarkable performance across various knowledge-intensive enterprise tasks. However, these models are primarily trained on unstructured, general knowledge, which limits their effectiveness in domain-specific applications-particularly when tasks involve structured data sources or sensitive enterprise information. We propose the first Structured Knowledge for Large Language Models Workshop - SKnow-LLM, which aims to bridge this gap by promoting research on innovative methodologies and practical applications in this area. Through keynote talks, panel discussions and paper presentations, the workshop will foster in-depth discussions on recent advances, identify existing challenges, and explore promising directions for integrating structured knowledge into LLMs.

The proposed ''SciSoc LLM Workshop: Large Language Models for Scientific and Societal Advances'' aims to explore the profound implications and potential of Large Language Models (LLMs) in driving forward scientific inquiry and addressing critical societal challenges. As LLMs such as GPT-4 continue to redefine boundaries in both complexity and capability, their integration into the scientific and societal domains is not just beneficial but essential. In particular, LLMs have demonstrated substantial value in improving our understanding of complex datasets and generating insights across various fields such as healthcare, environmental science, education, and public policy. By bringing together experts and enthusiasts from diverse fields, the workshop aims to foster a comprehensive understanding of how LLMs can redefine traditional research methodologies. Participants will explore innovative ways to harness the power of LLMs for greater efficiency and innovation in their respective fields, potentially catalyzing a new era of scientific and societal advancement.

We address the key challenge of size-induced distribution shifts in graph neural networks (GNNs) and their impact on the generalization of GNNs to larger graphs. Existing literature operates under diverse assumptions about distribution shifts, resulting in varying conclusions about the generalizability of GNNs. In contrast to prior work, we adopt a data-driven approach to identify and characterize the types of size-induced distribution shifts and explore their impact on GNN performance from a spectral standpoint, a perspective that has been largely underexplored. Leveraging the significant variance in graph sizes in real biological datasets, we analyze biological graphs and find that spectral differences-driven by subgraph patterns (e.g., average cycle length)-strongly correlate with GNN performance on larger, unseen graphs. Based on these insights, we propose three model-agnostic strategies to enhance GNNs' awareness of critical subgraph patterns, identifying size-intensive attention as the most effective approach. Extensive experiments with six GNN architectures and seven model-agnostic strategies across five datasets show that our size-intensive attention strategy significantly improves graph classification on test graphs 2 to 10 times larger than the training graphs, boosting F1 scores by up to 8% over strong baselines.

Collaborative filtering (CF) enables large-scale recommendation systems by encoding information from user-item interactions into dense ID-embedding tables. However, as embedding tables grow, closed-form solutions become impractical, necessitating the use of mini-batch gradient descent. Despite extensive work on loss function design, we argue that one core component of these pipelines is heavily overlooked: weight decay. Attaining high-performing models typically requires careful tuning of weight decay, regardless of loss, yet its necessity is not well understood. In this work, we question why weight decay is crucial in CF pipelines and how it impacts training. Through theoretical and empirical analysis, we uncover that weight decay's primary function is to encode popularity information into the magnitudes of the embedding vectors. Moreover, we find that tuning weight decay acts as a coarse, non-linear, knob to influence preference towards popular or unpopular items. Based on these findings, we propose PRISM (Popularity-awaRe Initialization Strategy for embedding Magnitudes), a straightforward yet effective solution to simplify the training of CF models. PRISM pre-encodes the popularity information typically learned through weight decay, eliminating its necessity. Our experiments show that PRISM improves performance by up to 4.77% and reduces training times by 38.48%, compared to state-of-the-art training strategies. Additionally, PRISM offers a cost-effective and meaningful strategy to mitigate popularity bias.

Random walk neural networks (RWNNs) have emerged as a promising approach for graph representation learning, leveraging recent advances in sequence models to process random walks. However, under realistic sampling constraints, RWNNs often fail to capture global structure even in small graphs due to incomplete node and edge coverage, limiting their expressivity. To address this, we propose \textit{random search neural networks} (RSNNs), which operate on random searches, each of which guarantees full node coverage. Theoretically, we demonstrate that in sparse graphs, only $O(\log |V|)$ searches are needed to achieve full edge coverage, substantially reducing sampling complexity compared to the $O(|V|)$ walks required by RWNNs (assuming walk lengths scale with graph size). Furthermore, when paired with universal sequence models, RSNNs are universal approximators. We lastly show RSNNs are probabilistically invariant to graph isomorphisms, ensuring their expectation is an isomorphism-invariant graph function. Empirically, RSNNs consistently outperform RWNNs on molecular and protein benchmarks, achieving comparable or superior performance with up to 16$\times$ fewer sampled sequences. Our work bridges theoretical and practical advances in random walk based approaches, offering an efficient and expressive framework for learning on sparse graphs.

Graph Neural Networks (GNNs) struggle to generalize when graphs exhibit both homophily (same-class connections) and heterophily (different-class connections). Specifically, GNNs tend to underperform for nodes with local homophily levels that differ significantly from the global homophily level. This issue poses a risk in user-centric applications where underrepresented homophily levels are present. Concurrently, fairness within GNNs has received substantial attention due to the potential amplification of biases via message passing. However, the connection between local homophily and fairness in GNNs remains underexplored. In this work, we move beyond global homophily and explore how local homophily levels can lead to unfair predictions. We begin by formalizing the challenge of fair predictions for underrepresented homophily levels as an out-of-distribution (OOD) problem. We then conduct a theoretical analysis that demonstrates how local homophily levels can alter predictions for differing sensitive attributes. We additionally introduce three new GNN fairness benchmarks, as well as a novel semi-synthetic graph generator, to empirically study the OOD problem. Across extensive analysis we find that two factors can promote unfairness: (a) OOD distance, and (b) heterophilous nodes situated in homophilous graphs. In cases where these two conditions are met, fairness drops by up to 24% on real world data, and 30% in semi-synthetic data. Collectively, our theoretical insights, empirical analysis, and algorithmic contributions unveil a previously overlooked source of unfairness rooted in the graph’s homophily information.

Collaborative Filtering (CF) methods dominate real-world recommender systems given their ability to learn high-quality, sparse ID-embedding tables that effectively capture user preferences. These tables scale linearly with the number of users and items, and are trained to ensure high similarity between embeddings of interacted user-item pairs, while maintaining low similarity for non-interacted pairs. Despite their high performance, encouraging dispersion for non-interacted pairs necessitates expensive regularization (e.g., negative sampling), hurting runtime and scalability. Existing research tends to address these challenges by simplifying the learning process, either by reducing model complexity or sampling data, trading performance for runtime. In this work, we move beyond model-level modifications and study the properties of the embedding tables under different learning strategies. Through theoretical analysis, we find that the singular values of the embedding tables are intrinsically linked to different CF loss functions. These findings are empirically validated on real-world datasets, demonstrating the practical benefits of higher stable rank -- a continuous version of matrix rank which encodes the distribution of singular values. Based on these insights, we propose an efficient warm-start strategy that regularizes the stable rank of the user and item embeddings. We show that stable rank regularization during early training phases can promote higher-quality embeddings, resulting in training speed improvements of up to 65.9%. Additionally, stable rank regularization can act as a proxy for negative sampling, allowing for performance gains of up to 21.2% over loss functions with small negative sampling ratios. Overall, our analysis unifies current CF methods under a new perspective -- their optimization of stable rank -- motivating a flexible regularization method that is easy to implement, yet effective at enhancing CF systems.

Distributed graph neural network (GNN) training facilitates learning on massive graphs that surpass the storage and computational capabilities of a single machine. Traditional distributed frameworks strive for performance parity with centralized training by maximally recovering cross-instance node dependencies, relying either on inter-instance communication or periodic fallback to centralized training. However, these processes create overhead and constrain the scalability of the framework. In this work, we propose a streamlined framework for distributed GNN training that eliminates these costly operations, yielding improved scalability, convergence speed, and performance over state-of-the-art approaches. Our framework (1) comprises independent trainers that asynchronously learn local models from locally available parts of the training graph and (2) synchronizes these local models only through periodic (time-based) model aggregation. Contrary to prevailing belief, our theoretical analysis shows that it is not essential to maximize the recovery of cross-instance node dependencies to achieve performance parity with centralized training. Instead, our framework leverages randomized assignment of nodes or super-nodes (i.e., collections of original nodes) to partition the training graph in order to enhance data uniformity and minimize discrepancies in gradient and loss function across instances. Experiments on social and e-commerce networks with up to 1.3 billion edges show that our proposed framework achieves state-of-the-art performance and 2.31 \(\times\) speedup compared to the fastest baseline despite using less training data.

Accurate confidence estimates are crucial for safe graph neural network (GNN) deployment, yet link prediction (LP) calibration is understudied. We provide novel insights into LP calibration by highlighting the importance of meaningful node-level uncertainties. In response, we propose E-ΔUQ, an architecture-agnostic framework leveraging stochastic centering to incorporate epistemic uncertainty into GNNs. Our work provides principles and three E-ΔUQ variants to improve trust in LP models, while introducing minimal overhead. Key results demonstrate properly handling node-level uncertainty improves edge calibration. We evaluate E-ΔUQ variants on citation networks and find that intermediate stochastic layers outperform alternatives by producing better node uncertainties. E-ΔUQ reduces calibration error by 15-50% and maintains comparable prediction fidelity.

While graph neural networks (GNNs) are widely used for node and graph representation learning tasks, the reliability of GNN uncertainty estimates under distribution shifts remains relatively under-explored. Indeed, while post-hoc calibration strategies can be used to improve in-distribution calibration, they need not also improve calibration under distribution shift. However, techniques which produce GNNs with better intrinsic uncertainty estimates are particularly valuable, as they can always be combined with post-hoc strategies later. Therefore, in this work, we propose G-$\Delta$UQ, a novel training framework designed to improve intrinsic GNN uncertainty estimates. Our framework adapts the principle of stochastic data centering to graph data through novel graph anchoring strategies, and is able to support partially stochastic GNNs. While, the prevalent wisdom is that fully stochastic networks are necessary to obtain reliable estimates, we find that the functional diversity induced by our anchoring strategies when sampling hypotheses renders this unnecessary and allows us to support G-$\Delta$UQ on pretrained models. Indeed, through extensive evaluation under covariate, concept and graph size shifts, we show that G-$\Delta$UQ leads to better calibrated GNNs for node and graph classification. Further, it also improves performance on the uncertainty-based tasks of out-of-distribution detection and generalization gap estimation. Overall, our work provides insights into uncertainty estimation for GNNs, and demonstrates the utility of G-$\Delta$UQ in obtaining reliable estimates.

Most real-world networks are noisy and incomplete samples from an unknown target distribution. Refining them by correcting corruptions or inferring unobserved regions typically improves downstream performance. Inspired by the impressive generative capabilities that have been used to correct corruptions in images, and the similarities between "in-painting" and filling in missing nodes and edges conditioned on the observed graph, we propose a novel graph generative framework, SGDM, which is based on subgraph diffusion. Our framework not only improves the scalability and fidelity of graph diffusion models, but also leverages the reverse process to perform novel, conditional generation tasks. In particular, through extensive empirical analysis and a set of novel metrics, we demonstrate that our proposed model effectively supports the following refinement tasks for partially observable networks: (T1) denoising extraneous subgraphs, (T2) expanding existing subgraphs and (T3) performing ``style" transfer by regenerating a particular subgraph to match the characteristics of a different node or subgraph.

Heterophily, or the tendency of connected nodes in networks to have different class labels or dissimilar features, has been identified as challenging for many Graph Neural Network (GNN) models. While the challenges of applying GNNs for node classification when class labels display strong heterophily are well understood, it is unclear how heterophily affects GNN performance in other important graph learning tasks where class labels are not available. In this work, we focus on the link prediction task and systematically analyze the impact of heterophily in node features on GNN performance. We first introduce formal definitions of homophilic and heterophilic link prediction tasks, and present a theoretical framework that highlights the different optimizations needed for the respective tasks. We then analyze how different link prediction encoders and decoders adapt to varying levels of feature homophily and introduce designs for improved performance. Based on our definitions, we identify and analyze six real-world benchmarks spanning from homophilic to heterophilic link prediction settings, with graphs containing up to 30M edges. Our empirical analysis on a variety of synthetic and real-world datasets confirms our theoretical insights and highlights the importance of adopting learnable decoders and GNN encoders with ego- and neighbor-embedding separation in message passing for link prediction tasks beyond homophily.

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.

While Graph Neural Networks (GNNs) are remarkably successful in a variety of high-impact applications, we demonstrate that, in link prediction, the common practices of including the edges being predicted in the graph at training and/or test have outsized impact on the performance of low-degree nodes. We theoretically and empirically investigate how these practices impact node-level performance across different degrees. Specifically, we explore three issues that arise: (I1) overfitting; (I2) distribution shift; and (I3) implicit test leakage. The former two issues lead to poor generalizability to the test data, while the latter leads to overestimation of the model's performance and directly impacts the deployment of GNNs. To address these issues in a systematic way, we introduce an effective and efficient GNN training framework, SpotTarget, which leverages our insight on low-degree nodes: (1) at training time, it excludes a (training) edge to be predicted if it is incident to at least one low-degree node; and (2) at test time, it excludes all test edges to be predicted (thus, mimicking real scenarios of using GNNs, where the test data is not included in the graph). SpotTarget helps researchers and practitioners adhere to best practices for learning from graph data, which are frequently overlooked even by the most widely-used frameworks. Our experiments on various real-world datasets show that SpotTarget makes GNNs up to 15× more accurate in sparse graphs, and significantly improves their performance for low-degree nodes in dense graphs.

How can we effectively detect fake reviews or fraudulent links on a website? How can we spot communities that suddenly appear based on users’ interactions? And how can we efficiently find the minimum cut in a large graph? All of these are related to the finding of dense subgraphs, a significant primitive problem in graph analysis with extensive applications across various domains. In this paper, we focus on formulating the problem of the densest subgraph detection and theoretically compare and contrast several correlated problems. Moreover, we propose a unified framework,GenDS, for the densest subgraph detection, provide some theoretical analysis based on the network flow and spectral graph theory, and devise simple and computationally efficient algorithms,SpecGDSandGepGDS, to solve it by leveraging the spectral properties and greedy search. We conduct thorough experiments on 40 real-world networks with up to 1.47 billion edges from various domains. We demonstrate that ourSpecGDSyields up to$58.6 \ \times$speedup and achieves better or approximately equal-quality solutions for the densest subgraph detection compared to the baselines.GepGDSalso reveals some properties of generalized eigenvalue problems for theGenDS. Also, our methods scale linearly with the graph size and are proven effective in applications such as finding collaborations that appear suddenly in an extensive, time-evolving co-authorship network.

Given a large graph, can we learn its node embeddings from a smaller summary graph? What is the relationship between embeddings learned from original graphs and their summary graphs? Graph representation learning plays an important role in many graph mining applications, but learning em-beddings of large-scale graphs remains a challenge. Recent works try to alleviate it via graph summarization, which typ-ically includes the three steps: reducing the graph size by combining nodes and edges into supernodes and superedges,learning the supernode embedding on the summary graph and then restoring the embeddings of the original nodes. How-ever, the justification behind those steps is still unknown. In this work, we propose GELSUMM, a well-formulated graph embedding learning framework based on graph sum-marization, in which we show the theoretical ground of learn-ing from summary graphs and the restoration with the three well-known graph embedding approaches in a closed form.Through extensive experiments on real-world datasets, we demonstrate that our methods can learn graph embeddings with matching or better performance on downstream tasks.This work provides theoretical analysis for learning node em-beddings via summarization and helps explain and under-stand the mechanism of the existing works.

Generalization error predictors (GEPs) aim to predict model performance on unseen distributions by deriving dataset-level error estimates from sample-level scores. However, GEPs often utilize disparate mechanisms (e.g., regressors, thresholding functions, calibration datasets, etc), to derive such error estimates, which can obfuscate the benefits of a particular scoring function. Therefore, in this work, we rigorously study the effectiveness of popular scoring functions (confidence, local manifold smoothness, model agreement), independent of mechanism choice. We find, absent complex mechanisms, that state-of-the-art confidence- and smoothness- based scores fail to outperform simple model-agreement scores when estimating error under distribution shifts and corruptions. Furthermore, on realistic settings where the training data has been compromised (e.g., label noise, measurement noise, under-sampling), we find that model-agreement scores continue to perform well and that ensemble diversity is important for improving its performance. Finally, to better understand the limitations of scoring functions, we demonstrate that simplicity bias, or the propensity of deep neural networks to rely upon simple but brittle features, can adversely affect GEP performance. Overall, our work carefully studies the effectiveness of popular scoring functions in realistic settings and helps to better understand their limitations.

Recent years have witnessed a surge of research interest in graph machine learning. However, the benchmark datasets available to the field are rather limited in both quantity and diversity, an issue particularly notable given the immense potential applications of graph learning. The lack of diverse benchmark datasets may have biased the development of graph machine learning techniques towards narrow directions. By crowdsourcing novel tasks and datasets, this workshop aims to increase the diversity of graph learning benchmarks, identify new demands of graph machine learning in general, and gain a better synergy of how concrete techniques perform on these benchmarks. Moreover, this workshop offers a platform for discussions of best practices in curating graph learning benchmarks and data-centric approaches for graph learning.

Brain graphs, which model the structural and functional relationships between brain regions, are crucial in neuroscientific and clinical applications that can be formulated as graph classification tasks. However, dense brain graphs pose computational challenges such as large time and memory consumption and poor model interpretability. In this paper, we investigate effective designs in Graph Neural Networks (GNNs) to sparsify brain graphs by eliminating noisy edges. Many prior works select noisy edges based on explainability or task-irrelevant properties, but this does not guarantee performance improvement when using the sparsified graphs. Additionally, the selection of noisy edges is often tailored to each individual graph, making it challenging to sparsify multiple graphs collectively using the same approach.

Network alignment, or the task of finding corresponding nodes in different networks, is an important problem formulation in many application domains. We propose CAPER, a multilevel alignment framework that Coarsens the input graphs, Aligns the coarsened graphs, Projects the alignment solution to finer levels and Refines the alignment solution. We show that CAPER can improve upon many different existing network alignment algorithms by enforcing alignment consistency across multiple graph resolutions: nodes matched at finer levels should also be matched at coarser levels. CAPER also accelerates the use of slower network alignment methods, at the modest cost of linear-time coarsening and refinement steps, by allowing them to be run on smaller coarsened versions of the input graphs. Experiments show that CAPER can improve upon diverse network alignment methods by an average of 33% in accuracy and/or an order of magnitude faster in runtime.

Understanding the training dynamics of deep neural networks (DNNs) is important as it can lead to improved training efficiency and task performance. Recent works have demonstrated that representing the wirings of neurons in feedforward DNNs as graphs is an effective strategy for understanding how architectural choices can affect performance. However, these approaches fail to model training dynamics since a single, static graph cannot capture how DNNs change over the course of training. Thus, in this work, we propose a compact, expressive temporal graph framework that effectively captures the dynamics of many workhorse architectures in computer vision. Specifically, our framework extracts an informative summary of graph properties (e.g., degree, eigenvector centrality) over a sequence of DNN graphs obtained during training. We demonstrate that the proposed framework captures useful dynamics by accurately predicting trained, task performance when using a summary over early training epochs (<5) across four different architectures and two image datasets. Moreover, by using a novel, highly-scalable DNN graph representation, we further demonstrate that the proposed framework captures generalizable dynamics as summaries extracted from smaller-width networks are effective when evaluated on larger widths.

In node classification tasks, graph convolutional neural networks (GCNs) have demonstrated competitive performance over traditional methods on diverse graph data. However, it is known that the performance of GCNs degrades with increasing number of layers (oversmoothing problem) and recent studies have also shown that GCNs may perform worse in heterophilous graphs, where neighboring nodes tend to belong to different classes (heterophily problem). These two problems are usually viewed as unrelated, and thus are studied independently, often at the graph filter level from a spectral perspective.We are the first to take a unified perspective to jointly explain the oversmoothing and heterophily problems at the node level. Specifically, we profile the nodes via two quantitative metrics: the relative degree of a node (compared to its neighbors) and the node-level heterophily. Our theory shows that the interplay of these two profiling metrics defines three cases of node behaviors, which explain the oversmoothing and heterophily problems jointly and can predict the performance of GCNs. Based on insights from our theory, we show theoretically and empirically the effectiveness of two strategies: structure-based edge correction, which learns corrected edge weights from structural properties (i.e., degrees), and feature-based edge correction, which learns signed edge weights from node features. Compared to other approaches, which tend to handle well either heterophily or oversmoothing, we show that our model, GGCN, which incorporates the two strategies performs well in both problems. We provide a longer version of this paper in [1] and codes on https://github.com/YujunYan/Heterophily_and_oversmoothing.

We bridge two research directions on graph neural networks (GNNs), by formalizing the relation between heterophily of node labels (i.e., connected nodes tend to have dissimilar labels) and the robustness of GNNs to adversarial attacks. Our theoretical and empirical analyses show that for homophilous graph data, impactful structural attacks always lead to reduced homophily, while for heterophilous graph data the change in the homophily level depends on the node degrees. These insights have practical implications for defending against attacks on real-world graphs: we deduce that separate aggregators for ego- and neighbor-embeddings, a design principle which has been identified to significantly improve prediction for heterophilous graph data, can also offer increased robustness to GNNs. Our comprehensive experiments show that GNNs merely adopting this design achieve improved empirical and certifiable robustness compared to the best-performing unvaccinated model. Additionally, combining this design with explicit defense mechanisms against adversarial attacks leads to an improved robustness with up to 18.33% performance increase under attacks compared to the best-performing vaccinated model.

Recent analyses of self-supervised learning (SSL) find the following data-centric properties to be critical for learning good representations: invariance to task-irrelevant semantics, separability of classes in some latent space, and recoverability of labels from augmented samples. However, given their discrete, non-Euclidean nature, graph datasets and graph SSL methods are unlikely to satisfy these properties. This raises the question: how do graph SSL methods, such as contrastive learning (CL), work well? To systematically probe this question, we perform a generalization analysis for CL when using generic graph augmentations (GGAs), with a focus on data-centric properties. Our analysis yields formal insights into the limitations of GGAs and the necessity of task-relevant augmentations. As we empirically show, GGAs do not induce task-relevant invariances on common benchmark datasets, leading to only marginal gains over naive, untrained baselines. Our theory motivates a synthetic data generation process that enables control over task-relevant information and boasts pre-defined optimal augmentations. This flexible benchmark helps us identify yet unrecognized limitations in advanced augmentation techniques (e.g., automated methods). Overall, our work rigorously contextualizes, both empirically and theoretically, the effects of data-centric properties on augmentation strategies and learning paradigms for graph SSL.

Temporal graph embedding has been widely studied thanks to its superiority in tasks such as prediction and recommendation. Despite the advances in algorithms and novel frameworks such as deep learning, there has been relatively little work on systematically studying the properties of temporal network models and their cornerstones, the graph time-series representations that are used in these approaches. This paper aims to fill this gap by introducing a general framework that extends an arbitrary existing static embedding approach to handle dynamic tasks, and conducting a systematic study of seven base static embedding methods and six temporal network models. Our framework generalizes static node embeddings derived from the time-series representation of stream data to the dynamic setting by modeling the temporal dependencies with classic models such as the reachability graph. While previous works on dynamic modeling and embedding have focused on representing a stream of timestamped edges using a time-series of graphs based on a specific time-scale (\eg, 1 month), we introduce the notion of an ε-graph time-series that uses a fixed number of edges for each graph, and show its superiority in practical settings over the standard solution. From the 42 methods that our framework subsumes, we find that leveraging the new ε-graph time-series representation and capturing temporal dependencies with the proposed reachability or summary graph tend to perform well. Furthermore, the new dynamic embedding methods based on our framework perform comparably and on average better than the state-of-the-art embedding methods designed specifically for temporal graphs in link prediction tasks.

Graph classification has a wide range of applications in bioinformatics, social sciences, automated fake news detection, web document classification, and more. In many practical scenarios, including web-scale applications, labels are scarce or hard to obtain. Unsupervised learning is thus a natural paradigm for these settings, but its performance often lags behind that of supervised learning. However, recently contrastive learning (CL) has enabled unsupervised computer vision models to perform comparably to supervised models. Theoretical and empirical works analyzing visual CL frameworks find that leveraging large datasets and task relevant augmentations is essential for CL framework success. Interestingly, graph CL frameworks report high performance while using orders of magnitude smaller data, and employing domain-agnostic graph augmentations (DAGAs) that can corrupt task relevant information.

While most network embedding techniques model the proximity between nodes in a network, recently there has been significant interest in structural embeddings that are based on node equivalences , a notion rooted in sociology: equivalences or positions are collections of nodes that have similar roles—i.e., similar functions, ties or interactions with nodes in other positions—irrespective of their distance or reachability in the network. Unlike the proximity-based methods that are rigorously evaluated in the literature, the evaluation of structural embeddings is less mature. It relies on small synthetic or real networks with labels that are not perfectly defined, and its connection to sociological equivalences has hitherto been vague and tenuous. With new node embedding methods being developed at a breakneck pace, proper evaluation, and systematic characterization of existing approaches will be essential to progress. To fill in this gap, we set out to understand what types of equivalences structural embeddings capture. We are the first to contribute rigorous intrinsic and extrinsic evaluation methodology for structural embeddings, along with carefully-designed, diverse datasets of varying sizes. We observe a number of different evaluation variables that can lead to different results (e.g., choice of similarity measure, classifier, and label definitions). We find that degree distributions within nodes’ local neighborhoods can lead to simple yet effective baselines in their own right and guide the future development of structural embedding. We hope that our findings can influence the design of further node embedding methods and also pave the way for more comprehensive and fair evaluation of structural embedding methods.

Graph Neural Networks (GNNs) have proven to be useful for many different practical applications. However, many existing GNN models have implicitly assumed homophily among the nodes connected in the graph, and therefore have largely overlooked the important setting of heterophily, where most connected nodes are from different classes. In this work, we propose a novel framework called CPGNN that generalizes GNNs for graphs with either homophily or heterophily. The proposed framework incorporates an interpretable compatibility matrix for modeling the heterophily or homophily level in the graph, which can be learned in an end-to-end fashion, enabling it to go beyond the assumption of strong homophily. Theoretically, we show that replacing the compatibility matrix in our framework with the identity (which represents pure homophily) reduces to GCN. Our extensive experiments demonstrate the effectiveness of our approach in more realistic and challenging experimental settings with significantly less training data compared to previous works: CPGNN variants achieve state-of-the-art results in heterophily settings with or without contextual node features, while maintaining comparable performance in homophily settings.

Relational knowledge bases (KBs) are commonly used to represent world knowledge in machines.However, while advantageous for their high degree of precision and interpretability, KBs are usually organized according to manually-defined schemas, which limit their expressiveness and require significant human efforts to engineer and maintain.In this review, we take a natural language processing perspective to these limitations, examining how they may be addressed in part by training deep contextual language models (LMs) to internalize and express relational knowledge in more flexible forms.We propose to organize knowledge representation strategies in LMs by the level of KB supervision provided, from no KB supervision at all to entity-and relation-level supervision.Our contributions are threefold:(1) We provide a high-level, extensible taxonomy for knowledge representation in LMs;(2) Within our taxonomy, we highlight notable models, evaluation tasks, and findings, in order to provide an up-to-date review of current knowledge representation capabilities in LMs; and(3) We suggest future research directions that build upon the complementary aspects of LMs and KBs as knowledge representations.

Codifying commonsense knowledge in machines is a longstanding goal of artificial intelligence.Recently, much progress toward this goal has been made with automatic knowledge base (KB) construction techniques.However, such techniques focus primarily on the acquisition of positive (true) KB statements, even though negative (false) statements are often also important for discriminative reasoning over commonsense KBs.As a first step toward the latter, this paper proposes NegatER, a framework that ranks potential negatives in commonsense KBs using a contextual language model (LM).Importantly, as most KBs do not contain negatives, NegatER relies only on the positive knowledge in the LM and does not require ground-truth negative examples.Experiments demonstrate that, compared to multiple contrastive data augmentation approaches, NegatER yields negatives that are more grammatical, coherent, and informative-leading to statistically significant accuracy improvements in a challenging KB completion task and confirming that the positive knowledge in LMs can be "repurposed" to generate negative knowledge.

Network alignment, or the task of finding meaningful node correspondences between nodes in di↵erent graphs, is an important graph mining task with many scientific and industrial applications.An important principle for network alignment is matched neighborhood consistency (MNC): nodes that are close in one graph should be matched to nodes that are close in the other graph.We theoretically demonstrate a close relationship between MNC and alignment accuracy.As many existing network alignment methods struggle to preserve topological consistency in di cult scenarios, we show how to refine their solutions by improving their MNC.Our refinement method, RefiNA, is straightforward to implement, admits scalable sparse approximation, and can be paired post hoc with any network alignment method.Extensive experiments show that RefiNA increases the accuracy of diverse unsupervised network alignment methods by up to 90%, making them robust enough to align graphs that are 5⇥ more topologically di↵erent than were considered in prior work.

While most network embedding techniques model the relative positions of nodes in a network, recently there has been significant interest in structural embeddings that model node role equivalences, irrespective of their distances to any specific nodes. We present PhUSION, a proximity-based unified framework for computing structural and positional node embeddings, which leverages well-established methods for calculating node proximity scores. Clarifying a point of contention in the literature, we show which step of PhUSION produces the different kinds of embeddings and what steps can be used by both. Moreover, by aggregating the PhUSION node embeddings, we obtain graph-level features that model information lost by previous graph feature learning and kernel methods. In a comprehensive empirical study with over 10 datasets, 4 tasks, and 35 methods, we systematically reveal successful design choices for node and graph-level machine learning with embeddings.

Multi-source entity linkage focuses on integrating knowledge from multiple sources by linking the records that represent the same real world entity. This is critical in high-impact applications such as data cleaning and user stitching. The state-of-the-art entity linkage pipelines mainly depend on supervised learning that requires abundant amounts of training data. However, collecting well-labeled training data becomes expensive when the data from many sources arrives incrementally over time. Moreover, the trained models can easily overfit to specific data sources, and thus fail to generalize to new sources due to significant differences in data and label distributions. To address these challenges, we present AdaMEL, a deep transfer learning framework that learns generic high-level knowledge to perform multi-source entity linkage. AdaMEL models the attribute importance that is used to match entities through an attribute-level self-attention mechanism, and leverages the massive unlabeled data from new data sources through domain adaptation to make it generic and data-source agnostic. In addition, AdaMEL is capable of incorporating an additional set of labeled data to more accurately integrate data sources with different attribute importance. Extensive experiments show that our framework achieves state-of-the-art results with 8.21% improvement on average over methods based on supervised learning. Besides, it is more stable in handling different sets of data sources in less runtime.

Our ability to generate, collect, and archive data related to everyday activities, such as interacting on social media, browsing the web, and monitoring well-being, is rapidly increasing. Getting the most benefit from this large-scale data requires analysis of patterns it contains, which is computationally intensive or even intractable. Summarization techniques produce compact data representations (summaries) that enable faster processing by complex algorithms and queries. This talk will cover summarization of interconnected data (graphs) [3], which can represent a variety of natural processes (e.g., friendships, communication). I will present an overview of my group's work on bridging the gap between research on summarized network representations and real-world problems. Examples include summarization of massive knowledge graphs for refinement [2] and on-device querying [4], summarization of graph streams for persistent activity detection [1], and summarization within graph neural networks for fast, interpretable classification [5]. I will conclude with open challenges and opportunities for future research.

Network alignment, the process of finding correspondences between nodes in different graphs, has many scientific and industrial applications. Existing unsupervised network alignment methods find suboptimal alignments that break up node neighborhoods, i.e. do not preserve matched neighborhood consistency. To improve this, we propose CONE-Align, which models intra-network proximity with node embeddings and uses them to match nodes across networks after aligning the embedding subspaces. Experiments on diverse, challenging datasets show that CONE-Align is robust and obtains 19.25% greater accuracy on average than the best-performing state-of-the-art graph alignment algorithm in highly noisy settings.

Network alignment is useful for multiple applications that require increasingly large graphs to be processed. Existing research approaches this as an optimization problem or computes the similarity based on node representations. However, the process of aligning every pair of nodes between relatively large networks is time-consuming and resource-intensive. In this paper, we propose a framework, called G-CREWE (Graph CompREssion With Embedding) to solve the network alignment problem. G-CREWE uses node embeddings to align the networks on two levels of resolution, a fine resolution given by the original network and a coarse resolution given by a compressed version, to achieve an efficient and effective network alignment. The framework first extracts node features and learns the node embedding via a Graph Convolutional Network (GCN). Then, node embedding helps to guide the process of graph compression and finally improve the alignment performance. As part of G-CREWE, we also propose a new compression mechanism called MERGE (Minimum DEgRee NeiGhbors ComprEssion) to reduce the size of the input networks while preserving the consistency in their topological structure. Experiments on all real networks show that our method is more than twice as fast as the most competitive existing methods while maintaining high accuracy.

The City of Detroit maintains an active fleet of over 2500 vehicles, spending an annual average of over $5 million on purchases and over $7.7 million on maintenance. Modeling patterns and trends in this data is of particular importance to a variety of stakeholders, particularly as Detroit emerges from Chapter 9 bankruptcy, but the structure in such data is complex, and the city lacks dedicated resources for in-depth analysis. The City of Detroit's Operations and Infrastructure Group and the University of Michigan initiated a collaboration which seeks to address this unmet need by analyzing data from the City of Detroit's vehicle fleet. This work presents a case study and provides the first data-driven benchmark, demonstrating a suite of methods to aid in data understanding and prediction for large vehicle maintenance datasets. We present analyses to address three key questions raised by the stakeholders, related to discovering multivariate maintenance patterns over time; predicting maintenance; and predicting vehicle- and fleet-level costs. We present a novel algorithm, PRISM, for automating multivariate sequential data analyses using tensor decomposition. This work is a first of its kind that presents both methodologies and insights to guide future civic data research.1

We present CODEX, a set of knowledge graph COmpletion Datasets EXtracted from Wikidata and Wikipedia that improve upon existing knowledge graph completion benchmarks in scope and level of difficulty.In terms of scope, CODEX comprises three knowledge graphs varying in size and structure, multilingual descriptions of entities and relations, and tens of thousands of hard negative triples that are plausible but verified to be false.To characterize CODEX, we contribute thorough empirical analyses and benchmarking experiments.First, we analyze each CODEX dataset in terms of logical relation patterns.Next, we report baseline link prediction and triple classification results on CODEX for five extensively tuned embedding models.Finally, we differentiate CODEX from the popular FB15K-237 knowledge graph completion dataset by showing that CODEX covers more diverse and interpretable content, and is a more difficult link prediction benchmark.Data, code, and pretrained models are available at https://bit.ly/2EPbrJs. |E| |R|Triples

Little is known about the trustworthiness of predictions made by knowledge graph embedding (KGE) models.In this paper we take initial steps toward this direction by investigating the calibration of KGE models, or the extent to which they output confidence scores that reflect the expected correctness of predicted knowledge graph triples.We first conduct an evaluation under the standard closed-world assumption (CWA), in which predicted triples not already in the knowledge graph are considered false, and show that existing calibration techniques are effective for KGE under this common but narrow assumption.Next, we introduce the more realistic but challenging open-world assumption (OWA), in which unobserved predictions are not considered true or false until ground-truth labels are obtained.Here, we show that existing calibration techniques are much less effective under the OWA than the CWA, and provide explanations for this discrepancy.Finally, to motivate the utility of calibration for KGE from a practitioner's perspective, we conduct a unique case study of human-AI collaboration, showing that calibrated predictions can improve human performance in a knowledge graph completion task.

The traditional setup of link prediction in networks assumes that a test set of node pairs, which is usually balanced, is available over which to predict the presence of links. However, in practice, there is no test set: the ground-truth is not known, so the number of possible pairs to predict over is quadratic in the number of nodes in the graph. Moreover, because graphs are sparse, most of these possible pairs will not be links. Thus, link prediction methods, which often rely on proximity-preserving embeddings or heuristic notions of node similarity, face a vast search space, with many pairs that are in close proximity, but that should not be linked. To mitigate this issue, we introduce LinkWaldo, a framework for choosing from this quadratic, massively-skewed search space of node pairs, a concise set of candidate pairs that, in addition to being in close proximity, also structurally resemble the observed edges. This allows it to ignore some high-proximity but low-resemblance pairs, and also identify high-resemblance, lower-proximity pairs. Our framework is built on a model that theoretically combines Stochastic Block Models (SBMs) with node proximity models. The block structure of the SBM maps out where in the search space new links are expected to fall, and the proximity identifies the most plausible links within these blocks, using locality sensitive hashing to avoid expensive exhaustive search. LinkWaldo can use any node representation learning or heuristic definition of proximity, and can generate candidate pairs for any link prediction method, allowing the representation power of current and future methods to be realized for link prediction in practice. We evaluate LinkWaldo on 13 networks across multiple domains, and show that on average it returns candidate sets containing 7-33% more missing and future links than both embedding-based and heuristic baselines' sets.

Frequent pattern mining is a key area of study that gives insights into the structure and dynamics of evolving networks, such as social or road networks. However, not only does a network evolve, but often the way that it evolves, itself evolves. Thus, knowing, in addition to patterns' frequencies, for how long and how regularly they have occurred-i.e., their persistence-can add to our understanding of evolving networks. In this work, we propose the problem of mining activity that persists through time in continually evolving networks-i.e., activity that repeatedly and consistently occurs. We extend the notion of temporal motifs to capture activity among specific nodes, in what we call activity snippets, which are small sequences of edge-updates that reoccur. We propose axioms and properties that a measure of persistence should satisfy, and develop such a persistence measure. We also propose PENminer, an efficient framework for mining activity snippets' Persistence in Evolving Networks, and design both offline and streaming algorithms. We apply PENminer to numerous real, large-scale evolving networks and edge streams, and find activity that is surprisingly regular over a long period of time, but too infrequent to be discovered by aggregate count alone, and bursts of activity exposed by their lack of persistence. Our findings with PENminer include neighborhoods in NYC where taxi traffic persisted through Hurricane Sandy, the opening of new bike-stations, characteristics of social network users, and more. Moreover, we use PENminer towards identifying anomalies in multiple networks, outperforming baselines at identifying subtle anomalies by 9.8-48% in AUC.

A significant effort has been made to train neural networks that replicate algorithmic reasoning, but they often fail to learn the abstract concepts underlying these algorithms. This is evidenced by their inability to generalize to data distributions that are outside of their restricted training sets, namely larger inputs and unseen data. We study these generalization issues at the level of numerical subroutines that comprise common algorithms like sorting, shortest paths, and minimum spanning trees. First, we observe that transformer-based sequence-to-sequence models can learn subroutines like sorting a list of numbers, but their performance rapidly degrades as the length of lists grows beyond those found in the training set. We demonstrate that this is due to attention weights that lose fidelity with longer sequences, particularly when the input numbers are numerically similar. To address the issue, we propose a learned conditional masking mechanism, which enables the model to strongly generalize far outside of its training range with near-perfect accuracy on a variety of algorithms. Second, to generalize to unseen data, we show that encoding numbers with a binary representation leads to embeddings with rich structure once trained on downstream tasks like addition or multiplication. This allows the embedding to handle missing data by faithfully interpolating numbers not seen during training.

We investigate the representation power of graph neural networks in the semi-supervised node classification task under heterophily or low homophily, i.e., in networks where connected nodes may have different class labels and dissimilar features. Many popular GNNs fail to generalize to this setting, and are even outperformed by models that ignore the graph structure (e.g., multilayer perceptrons). Motivated by this limitation, we identify a set of key designs—ego- and neighbor-embedding separation, higher-order neighborhoods, and combination of intermediate representations—that boost learning from the graph structure under heterophily. We combine them into a graph neural network, H2GCN, which we use as the base method to empirically evaluate the effectiveness of the identified designs. Going beyond the traditional benchmarks with strong homophily, our empirical analysis shows that the identified designs increase the accuracy of GNNs by up to 40% and 27% over models without them on synthetic and real networks with heterophily, respectively, and yield competitive performance under homophily.

Individuals' personal information collections (their emails, files, appointments, web searches, contacts, etc) offer a wealth of insights into the organization and structure of their everyday lives. In this paper we address the task of learning representations of personal information items to capture individuals' ongoing activities, such as projects and tasks: Such representations can be used in activity-centric applications like personal assistants, email clients, and productivity tools to help people better manage their data and time. We propose a graph-based approach that leverages the inherent interconnected structure of personal information collections, and derive efficient, exact techniques to incrementally update representations as new data arrive. We demonstrate the strengths of our graph-based representations against competitive baselines in a novel intrinsic rating task and an extrinsic recommendation task.

Knowledge graphs (KGs) store highly heterogeneous information about the world in the structure of a graph, and are useful for tasks such as question answering and reasoning. However, they often contain errors and are missing information. Vibrant research in KG refinement has worked to resolve these issues, tailoring techniques to either detect specific types of errors or complete a KG.

OBJECTIVE: In applying machine learning (ML) to electronic health record (EHR) data, many decisions must be made before any ML is applied; such preprocessing requires substantial effort and can be labor-intensive. As the role of ML in health care grows, there is an increasing need for systematic and reproducible preprocessing techniques for EHR data. Thus, we developed FIDDLE (Flexible Data-Driven Pipeline), an open-source framework that streamlines the preprocessing of data extracted from the EHR. MATERIALS AND METHODS: Largely data-driven, FIDDLE systematically transforms structured EHR data into feature vectors, limiting the number of decisions a user must make while incorporating good practices from the literature. To demonstrate its utility and flexibility, we conducted a proof-of-concept experiment in which we applied FIDDLE to 2 publicly available EHR data sets collected from intensive care units: MIMIC-III and the eICU Collaborative Research Database. We trained different ML models to predict 3 clinically important outcomes: in-hospital mortality, acute respiratory failure, and shock. We evaluated models using the area under the receiver operating characteristics curve (AUROC), and compared it to several baselines. RESULTS: Across tasks, FIDDLE extracted 2,528 to 7,403 features from MIMIC-III and eICU, respectively. On all tasks, FIDDLE-based models achieved good discriminative performance, with AUROCs of 0.757-0.886, comparable to the performance of MIMIC-Extract, a preprocessing pipeline designed specifically for MIMIC-III. Furthermore, our results showed that FIDDLE is generalizable across different prediction times, ML algorithms, and data sets, while being relatively robust to different settings of user-defined arguments. CONCLUSIONS: FIDDLE, an open-source preprocessing pipeline, facilitates applying ML to structured EHR data. By accelerating and standardizing labor-intensive preprocessing, FIDDLE can help stimulate progress in building clinically useful ML tools for EHR data.

Structural roles define sets of structurally similar nodes that are more similar to nodes inside the set than outside, whereas communities define sets of nodes with more connections inside the set than outside. Roles based on structural similarity and communities based on proximity are fundamentally different but important complementary notions. Recently, the notion of structural roles has become increasingly important and has gained a lot of attention due to the proliferation of work on learning representations (node/edge embeddings) from graphs that preserve the notion of roles. Unfortunately, recent work has sometimes confused the notion of structural roles and communities (based on proximity) leading to misleading or incorrect claims about the capabilities of network embedding methods. As such, this article seeks to clarify the misconceptions and key differences between structural roles and communities, and formalize the general mechanisms (e.g., random walks and feature diffusion) that give rise to community- or role-based structural embeddings. We theoretically prove that embedding methods based on these mechanisms result in either community- or role-based structural embeddings. These mechanisms are typically easy to identify and can help researchers quickly determine whether a method preserves community- or role-based embeddings. Furthermore, they also serve as a basis for developing new and improved methods for community- or role-based structural embeddings. Finally, we analyze and discuss applications and data characteristics where community- or role-based embeddings are most appropriate.

For trajectory data that tend to have beyond first-order (i.e., non-Markovian) dependencies, higher-order networks have been shown to accurately capture details lost with the standard aggregate network representation. At the same time, representation learning has shown success on a wide range of network tasks, removing the need to hand-craft features for these tasks. In this work, we propose a node representation learning framework called EVO or Embedding Variable Orders, which captures non-Markovian dependencies by combining work on higher-order networks with work on node embeddings. We show that EVO outperforms baselines in tasks where high-order dependencies are likely to matter, demonstrating the benefits of considering high-order dependencies in node embeddings. We also provide insights into when it does or does not help to capture these dependencies. To the best of our knowledge, this is the first work on representation learning for higher-order networks.

Facilitating audience participation in a music performance brings with it challenges in involving non-expert users in large-scale collaboration. A musical piece needs to be created live, over a short period of time, with limited communication channels. To address this challenge, we propose to incorporate social interaction through mobile music instruments that the audience is given to play with, and examine how this feature sustains and affects the audience involvement. We test this idea with an audience participation music system, Crowd in C. We realized a participation-based musical performance with the system and validated our approach by analyzing the interaction traces of the audience at a performance. The result indicates that the audience members were actively engaged throughout the performance, with multiple layers of social interaction available in the system. We also present how the social interactivity among the audience shaped their interaction in the music making process.

Graph classification is an important problem in many fields, from bioinformatics and neuroscience to computer vision and social network analysis. That said, the task of comparing graphs for the purpose of graph classification faces several major challenges. In particular, an effective graph comparison method must (1) expressively and inductively compare graphs; (2) efficiently compare large graphs; and (3) enable the use of fast machine learning models for graph classification. To address such challenges, we propose Randomized Grid Mapping (RGM), a fast-to-compute feature map that represents a graph via the distribution of its node embeddings in feature space. We justify RGM with close connections to kernel methods: RGM provably approximates the Laplacian kernel mean map and has the multiresolution properties of the pyramid match kernel. We also show that RGM can be extended to incorporate node labels using the Weisfeiler-Lehman framework. Extensive experiments show that graph classification accuracy with RGM feature maps is better than or competitive with many powerful graph kernels, unsupervised graph feature mappings, and deep neural networks. Moreover, comparing graphs based on their node embeddings with RGM is up to an order of magnitude faster than competitive baselines, while maintaining high classification accuracy.

The increasing scale of encyclopedic knowledge graphs (KGs) calls for summarization as a way to help users efficiently access and distill world knowledge. Motivated by the disparity between individuals' limited information needs and the massive scale of KGs, in this paper we propose a new problem called personalized knowledge graph summarization. The goal is to construct compact "personal summaries" of KGs containing only the facts most relevant to individuals' interests. Such summaries can be stored and utilized on-device, allowing individuals private, anytime access to the information that interests them most. We formalize the problem as one of constructing a sparse graph, or summary, that maximizes a user's inferred "utility" over a given KG, subject to a user-and device-specific constraint on the summary's size. To solve it, we propose GLIMPSE, a summarization framework that provides theoretical guarantees on the summary's utility and is linear in the number of edges in the KG. In an evaluation with real user queries to open-source, encyclopedic KGs of up to one billion triples, we show that GLIMPSE efficiently creates summaries that outperform strong baselines by up to 19% in query answering F1 score.

Email is ubiquitous in the workplace. Naturally, machine learning models that make third-party email clients "smarter" can dramatically impact employees' productivity and efficiency. Motivated by this potential, we study the task of professional role inference from email data, which is crucial for email prioritization and contact recommendation systems. The central question we address is: Given limited data about employees, as is common in third-party email applications, can we infer where in the organizational hierarchy these employees belong based on their email behavior? Toward our goal, in this paper we study professional role inference on a unique new email dataset comprising billions of email exchanges across thousands of organizations. Taking a network approach in which nodes are employees and edges represent email communication, we propose EMBER, or EMBedding Email-based Roles, which finds email-centric embeddings of network nodes to be used in professional role inference tasks. EMBER automatically captures behavioral similarity between employees in the email network, leading to embeddings that naturally distinguish employees of different hierarchical roles. EMBER often outperforms the state-of-the-art by 2-20% in role inference accuracy and 2.5-344x in speed. We also use EMBER with our unique dataset to study how inferred professional roles compare between organizations of different sizes and sectors, gaining new insights into organizational hierarchy.

Motivated by the computational and storage challenges that dense embeddings pose, we introduce the problem of latent network summarization that aims to learn a compact, latent representation of the graph structure with dimensionality that is independent of the input graph size (\i.e., #nodes and #edges), while retaining the ability to derive node representations on the fly. We propose Multi-LENS, an inductive multi-level latent network summarization approach that leverages a set of relational operators and relational functions (compositions of operators) to capture the structure of egonets and higher-order subgraphs, respectively. The structure is stored in low-rank, size-independent structural feature matrices, which along with the relational functions comprise our latent network summary. Multi-LENS is general and naturally supports both homogeneous and heterogeneous graphs with or without directionality, weights, attributes or labels. Extensive experiments on real graphs show 3.5-34.3% improvement in AUC for link prediction, while requiring 80-2152x less output storage space than baseline embedding methods on large datasets. As application areas, we show the effectiveness of Multi-LENS in detecting anomalies and events in the Enron email communication graph and Twitter co-mention graph.

Mapping the human brain, or understanding how certain brain regions relate to specific aspects of cognition, has been and remains an active area of neuroscience research. Functional magnetic resonance imaging (fMRI) data---in the form of images, time series or graphs---are central in this research, but pose many challenges in phenotype prediction tasks (e.g., noisy, small training samples). Standardly employed handcrafted models and newly proposed neural network methods pose limitations in the expressive power and interpretability, respectively, in this context. In this work focusing on fMRI-derived brain graphs, a modality that partially handles some challenges of fMRI data, we propose a grouping-based interpretable neural network model, GroupINN, that effectively classifies cognitive performance with 85% fewer model parameters than baseline deep models, while also identifying the most predictive brain subnetworks within several task-specific contexts. Our method incorporates the idea of node grouping into the design of the neural network. That way, unlike other methods that employ clustering as a preprocessing step to reorder nodes, GroupINN learns the node grouping and extracts graph features jointly. Experiments on task-based fMRI datasets show that our method is $2.6-69\times$ faster than other deep models, while achieving comparable or better accuracy and providing interpretability.

Motivated by the problem of human-trafficking, where it is often observed that criminal organizations are linked and behave similarly over time, we introduce the problem of Coupled Clustering of Time-series and their underlying Network. The goal is to find tightly connected subgroups of nodes that also have similar node-specific time series (temporal—not necessarily structural—behavior). We formulate the problem as a coupled matrix factorization for the time series, combined with regularization for network smoothness. We propose CCTN, and an incrementally-updated counterpart, CCTN-inc, which efficiently handles network updates. Extensive experiments show that CCTN is up to 4x more accurate than baselines that consider graph structure or time series alone, and CCTN-inc is up to 55x faster than CCTN. As an application, we explore an exclusive database with millions of online ads on human trafficking, and successfully deploy our technique to detect criminal organizations.

Graph representations have increasingly grown in popularity during the last years. Existing embedding approaches explicitly encode network structure. Despite their good performance in downstream processes (e.g., node classification), there is still room for improvement in different aspects, like effectiveness. In this paper, we propose, t-PNE, a method that addresses this limitation. Contrary to baseline methods, which generally learn explicit node representations by solely using an adjacency matrix, t-PNE avails a multi-view information graph-the adjacency matrix represents the first view, and a nearest neighbor adjacency, computed over the node features, is the second view-in order to learn explicit and implicit node representations, using the Canonical Polyadic (a.k.a. CP) decomposition. We argue that the implicit and the explicit mapping from a higher-dimensional to a lower-dimensional vector space is the key to learn more useful and highly predictable representations. Extensive experiments show that t-PNE drastically outperforms baseline methods by up to 158.6% with respect to Micro-Fl, in several multi-label classification problems.

Problems involving multiple networks are prevalent in many scientific and other domains. In particular, network alignment, or the task of identifying corresponding nodes in different networks, has applications across the social and natural sciences. Motivated by recent advancements in node representation learning for single-graph tasks, we propose REGAL (REpresentation learning-based Graph ALignment), a framework that leverages the power of automatically-learned node representations to match nodes across different graphs. Within REGAL we devise xNetMF, an elegant and principled node embedding formulation that uniquely generalizes to multi-network problems. Our results demonstrate the utility and promise of unsupervised representation learning-based network alignment in terms of both speed and accuracy. REGAL runs up to 30x faster in the representation learning stage than comparable methods, outperforms existing network alignment methods by 20 to 30% accuracy on average, and scales to networks with millions of nodes each.

Recent advances in computing resources have made it possible to collect enormous amounts of interconnected data, such as social media interactions, web activity, knowledge bases, product and service purchases, autonomous vehicle routing, smart home sensor data, and more. The massive scale and complexity of this data, however, not only vastly surpasses human processing power, but also goes beyond limitations with regard to computation and storage. That is, there is an urgent need for methods and tools that summarize large interconnected data to enable faster computations, storage reduction, interactive large-scale visualization and understanding, and pattern discovery. Network summarization-which aims to find a small representation of an original, larger graph-features a variety of methods with different goals and for different input data representations (e.g., attributed graphs, time-evolving or streaming graphs, heterogeneous graphs). The objective of this tutorial is to give a systematic overview of methods for summarizing and explaining graphs at different scales: the node-group level, the network level, and the multi-network level. We emphasize the current challenges, present real-world applications, and highlight the open research problems in this vibrant research area.

From artificial intelligence to network security to hardware design, it is well-known that computing research drives many important technological and societal advancements. However, less is known about the long-term career paths of the people behind these innovations. What do their careers reveal about the evolution of computing research? Which institutions were and are the most important in this field, and for what reasons? Can insights into computing career trajectories help predict employer retention?

As graph datasets grow, faster data mining methods become indispensable. Random Walk with Restart (RWR), belief propagation, semi-supervised learning, and more graph methods can be expressed as a set of linear equations. In this work, we focus on solving such equations fast and accurately when large number of queries need to be handled. We use RWR as a case study, since it is widely used not only to evaluate the importance of a node, but also as a basis for more complex tasks, e.g., representation learning and community detection. We introduce a new, intuitive two-step divide-and-conquer formulation and a corresponding parallelizable method, FlowR, for solving RWR with two goals: (i) fast and accurate computation under multiple queries; (ii) one-time message exchange between subproblems. We further speed up our proposed method by extending our formulation to carefully designed overlapping subproblems (FlowR-OV) and by leveraging the strengths of iterative methods (FlowR-Hyb). Extensive experiments on synthetic and real networks with up to ∼8 million edges show that our methods are accurate and outperform in runtime various state-of-the-art approaches, running up to 34× faster in preprocessing and up to 32× faster in query time.

Data integration is frequently required to obtain the full value of data from multiple sources. In spite of extensive research on tools to assist users, data integration remains hard, particularly for users with limited technical proficiency. To address this barrier, we study how much we can do with no user guidance. Our vision is that the user should merely specify two input datasets to be joined and get a meaningful integrated result. It turns out that our vision can be realized if the system can correctly determine the join key, for example based on domain knowledge.

Viruses and bacteria are critical components of the human microbiome and play important roles in health and disease. Most previous work has relied on studying bacteria and viruses independently, thereby reducing them to two separate communities. Such approaches are unable to capture how these microbial communities interact, such as through processes that maintain community robustness or allow phage-host populations to co-evolve. We implemented a network-based analytical approach to describe phage-bacteria network diversity throughout the human body. We built these community networks using a machine learning algorithm to predict which phages could infect which bacteria in a given microbiome. Our algorithm was applied to paired viral and bacterial metagenomic sequence sets from three previously published human cohorts. We organized the predicted interactions into networks that allowed us to evaluate phage-bacteria connectedness across the human body. We observed evidence that gut and skin network structures were person-specific and not conserved among cohabitating family members. High-fat diets appeared to be associated with less connected networks. Network structure differed between skin sites, with those exposed to the external environment being less connected and likely more susceptible to network degradation by microbial extinction events. This study quantified and contrasted the diversity of virome-microbiome networks across the human body and illustrated how environmental factors may influence phage-bacteria interactive dynamics. This work provides a baseline for future studies to better understand system perturbations, such as disease states, through ecological networks.

Given the set of social interactions of a user, how can we detect changes in interaction patterns over time? While most previous work has focused on studying network-wide properties and spotting outlier users, the dynamics of individual user interactions remain largely unexplored. This work sets out to explore those dynamics in a way that is minimally invasive to privacy, thus, avoids to rely on the textual content of user posts---except for validation. Our contributions are two-fold. First, in contrast to previous studies, we challenge the use of a fixed interval of observation. We introduce and empirically validate the "Temporal Asymmetry Hypothesis", which states that appropriate observation intervals should vary both among users and over time for the same user. We validate this hypothesis using eight different datasets, including email, messaging, and social networks data. Second, we propose iNET, a comprehensive analytic and visualization framework which provides personalized insights into user behavior and operates in a streaming fashion. iNET learns personalized baseline behaviors of users and uses them to identify events that signify changes in user behavior. We evaluate the effectiveness of iNET by analyzing more than half a million interactions from Facebook users. Labeling of the identified changes in user behavior showed that iNET is able to capture a wide spectrum of exogenous and endogenous events, while the baselines are less diverse in nature and capture only 66% of that spectrum. Furthermore, iNET exhibited the highest precision (95%) compared to all competing approaches.

We propose BotWalk, a near-real time adaptive Twitter exploration algorithm to identify bots exhibiting novel behavior. Due to suspension pressure, Twitter bots are constantly changing their behavior to evade detection. Traditional supervised approaches to bot detection are non-adaptive and thus cannot identify novel bot behaviors. We therefore devise an unsupervised approach, which allows us to identify bots as they evolve. We characterize users with a behavioral feature vector which consists of (well-studied in isolation) metadata-, content-, temporal-, and network-based features. We identify a random bot from our seed bank, populated initially by previously-labeled bots, gather this user's followers' features from Twitter in real time, and employ an unsupervised ensemble anomaly detection method in the multi-dimensional behavioral space. These potential bots are folded into the seed bank and the process is then repeated, with the new seeds' features allowing us to adaptively identify novel bot behavior. BotWalk allows for the identification of on average 6,000 potential bots a day. Our method allowed us to detect 7,995 previously undiscovered bots from a sample of 15 seed bots with a precision of 90%.

Given the soaring amount of data being generated daily, graph mining tasks are becoming increasingly challenging, leading to tremendous demand for summarization techniques. Feature selection is a representative approach that simplifies a dataset by choosing features that are relevant to a specific task, such as classification, prediction, and anomaly detection. Although it can be viewed as a way to summarize a graph in terms of a few features, it is not well-defined for exploratory analysis, and it operates on a set of observations jointly rather than conditionally (i.e., feature selection from many graphs vs. selection for an input graph conditioned on other graphs). In this work, we introduce EAGLE (Exploratory Analysis of Graphs with domain knowLEdge), a novel method that creates interpretable, feature-based, and domain-specific graph summaries in a fully automatic way. That is, the same graph in different domains-e.g., social science and neuroscience-will be described via different EAGLE summaries, which automatically leverage the domain knowledge and expectations. We propose an optimization formulation that seeks to find an interpretable summary with the most representative features for the input graph so that it is: diverse, concise, domain-specific, and efficient. Extensive experiments on synthetic and real-world datasets with up to ~1M edges and ~400 features demonstrate the effectiveness and efficiency of EAGLE and its benefits over existing methods. We also show how our method can be applied to various graph mining tasks, such as classification and exploratory analysis.

Discovering and analyzing networks from non-network data is a task with applications in fields as diverse as neuroscience, genomics, energy, economics, and more. In these domains, networks are often constructed out of multiple time series by computing measures of association or similarity between pairs of series. The nodes in a discovered graph correspond to time series, which are linked via edges weighted by the association scores of their endpoints. After graph construction, the network may be thresholded such that only the edges with stronger weights remain and the desired sparsity level is achieved. While this approach is feasible for small datasets, its quadratic time complexity does not scale as the individual time series length and the number of compared series increase. Thus, to avoid the costly step of building a fully-connected graph before sparsification, we propose a fast network discovery approach based on probabilistic hashing of randomly selected time series subsequences. Evaluation on real data shows that our methods construct graphs nearly 15 times as fast as baseline methods, while achieving both network structure and accuracy comparable to baselines in task-based evaluation.

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.

Given a directed graph, how should we label both its outgoing and incoming edges to achieve better disk locality and support neighborhood-related edge queries? In this paper, we answer this question with edge-labeling schemes GrdRandom and FlipInOut, to label edges with integers based on the premise that edges should be assigned integer identifiers exploiting their consecutiveness to a maximum degree.

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 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.

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.

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.

It has been suggested that online search and retrieval contributes to the intellectual isolation of users within their preexisting ideologies, where people's prior views are strengthened and alternative viewpoints are infrequently encountered. This so-called "filter bubble" phenomenon has been called out as especially detrimental when it comes to dialog among people on controversial, emotionally charged topics, such as the labeling of genetically modified food, the right to bear arms, the death penalty, and online privacy. We seek to identify and study information-seeking behavior and access to alternative versus reinforcing viewpoints following shocking, emotional, and large-scale news events. We choose for a case study to analyze search and browsing on gun control/rights, a strongly polarizing topic for both citizens and leaders of the United States. We study the period of time preceding and following a mass shooting to understand how its occurrence, follow-on discussions, and debate may have been linked to changes in the patterns of searching and browsing. We employ information-theoretic measures to quantify the diversity of Web domains of interest to users and understand the browsing patterns of users. We use these measures to characterize the influence of news events on these web search and browsing patterns.

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.

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.

Behavioral researchers spend considerable amount of time coding video data to systematically extract meaning from subtle human actions and emotions. In this paper, we present Glance, a tool that allows researchers to rapidly query, sample, and analyze large video datasets for behavioral events that are hard to detect automatically. Glance takes advantage of the parallelism available in paid online crowds to interpret natural language queries and then aggregates responses in a summary view of the video data. Glance provides analysts with rapid responses when initially exploring a dataset, and reliable codings when refining an analysis. Our experiments show that Glance can code nearly 50 minutes of video in 5 minutes by recruiting over 60 workers simultaneously, and can get initial feedback to analysts in under 10 seconds for most clips. We present and compare new methods for accurately aggregating the input of multiple workers marking the spans of events in video data, and for measuring the quality of their coding in real-time before a baseline is established by measuring the variance between workers. Glance's rapid responses to natural language queries, feedback regarding question ambiguity and anomalies in the data, and ability to build on prior context in followup queries allow users to have a conversation-like interaction with their data - opening up new possibilities for naturally exploring video data.

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.

How can we find the virtual twin (i.e., the same or similar user) on Linked In for a user on Facebook? How can we effectively link an information network with a social network to support cross-network search? Graph alignment - the task of finding the node correspondences between two given graphs - is a fundamental building block in numerous application domains, such as social networks analysis, bioinformatics, chemistry, pattern recognition. In this work, we focus on aligning bipartite graphs, a problem which has been largely ignored by the extensive existing work on graph matching, despite the ubiquity of those graphs (e.g., users-groups network). We introduce a new optimization formulation and propose an effective and fast algorithm to solve it. We also propose a fast generalization of our approach to align unipartite graphs. The extensive experimental evaluations show that our method outperforms the state-of-art graph matching algorithms in both alignment accuracy and running time, being up to 10x more accurate or 174x faster on real graphs.

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.

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.

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.

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.