Hypergraphs, in contrast to general graphs, utilize hyperedges to connect multiple nodes, thereby inherently facilitating the representation of higher-order relational structures. To leverage the benefits of hypergraphs, several Hypergraph Neural Networks (HyGNNs) have been proposed to model hypergraph structures. Although existing HyGNNs excel at capturing complex relationships in homophilic hypergraphs, they still face challenges in modeling heterophilic hypergraphs, as most existing HyGNNs are designed based on the homophily principle. Recent studies have attempted to leverage attention mechanisms that are less reliant on the homophily principle. However, these attention mechanisms remain ineffective for nodes in heterophilic hypergraphs. To tackle the aforementioned challenges, we propose a novel Broadcast HyperGraph Neural Network (BHyGNN) to adaptively broadcast node information to learn more effective node representations in heterophilic hypergraphs. Specifically, we devise a novel Variational Broadcast Autoencoder Network to sample the broadcast and receive actions to propagate information between nodes and hyperedges. Moreover, we design an incorporation transformer mechanism to perform the estimated broadcast or receive actions to learn the hyperedge or node representations, incorporating the information from both sides. Extensive experiments over five benchmark heterophilic hypergraph datasets and six homophilic hypergraph datasets demonstrate the effectiveness of BHyGNN over all baseline methods. Our source code and datasets are available at https://github.com/Tianyi-Billy-Ma/BHyGNN.

Graph learning models have demonstrated great prowess in learning expressive representations from large-scale graph data in a wide variety of real-world scenarios. As a prevalent strategy for training powerful graph learning models, the ''pre-training, adaptation'' scheme first pre-trains graph learning models on unlabeled graph data in a self-supervised manner and then adapts them to specific downstream tasks. During the adaptation phase, graph prompting emerges as a promising approach that learns trainable prompts while keeping the pre-trained graph learning models unchanged. In this paper, we present a systematic review of recent advancements in graph prompting. First, we introduce representative graph pre-training methods that serve as the foundation step of graph prompting. Next, we review mainstream techniques in graph prompting and elaborate on how they design learnable prompts for graph prompting. Furthermore, we summarize the real-world applications of graph prompting from different domains. Finally, we discuss several open challenges in existing studies with promising future directions in this field.

Foundation models have revolutionized machine learning by enabling general-purpose reasoning across diverse tasks and domains. These models, pretrained on large-scale data, demonstrate strong adaptability with minimal task-specific supervision, leading to breakthroughs in natural language processing and computer vision. Inspired by this paradigm, Graph Foundation Models (GFMs) have emerged to extend the benefits of foundation models to graph-structured data, which is pretrained on massive graphs and can be fast adapted to different downstream tasks. In this paper, we provide a comprehensive survey of the state-of-the-art techniques of graph foundation models. In particular, we (1) formally categorize the challenges in designing graph foundation models; (2) comprehensively review the existing and recent advances of graph foundation models; (3) extend the graph foundation models in real-world problems; and (4) elucidate open questions and future research directions. Our systematic review summarizes representative models, highlights key design principles, and provides comparative analyses. This survey introduces major topics within foundation models and offers a guide to a new frontier of graph learning. Our extended survey is available at https://arxiv.org/abs/2505.15116.

Modern machine learning techniques, particularly deep learning, have showcased remarkable efficacy across numerous knowledge discovery and data mining applications. However, the advancement of these methods is frequently impeded by resource constraint challenges in many scenarios, such as limited labeled data (data-level), small model size requirements in real-world computing platforms (model-level), and efficient mapping of the computations to heterogeneous target hardware (system-level). Addressing all these factors is crucial for effectively and efficiently deploying developed models across a broad spectrum of real-world systems, including large-scale social network analysis, recommendation systems, and real-time anomaly detection. Therefore, there is a critical need to develop efficient learning techniques to address the challenges posed by resource limitations, whether from data, model/algorithm, or system/hardware perspectives. The proposed third international workshop on ''Resource-Efficient Learning for Knowledge Discovery (RelKD 2025)'' will provide a great venue for academic researchers and industrial practitioners to share challenges, solutions, and future opportunities for resource-efficient learning.

The prevalence of unhealthy eating habits has become a growing concern in the United States. However, popular food recommendation platforms, such as Yelp, tend to prioritize users' dietary preferences over the healthiness of their choices. While some efforts have focused on developing health-aware food recommendation systems, personalization based on specific health conditions remains underexplored. Additionally, the lack of interpretability in these systems prevents users from evaluating the reliability of recommendations, limiting their practical adoption. To address these issues, we introduce two large-scale personalized health-aware food recommendation benchmarks at the first attempt. Building on this, we propose a novel framework called the Multi-Objective Personalized Interpretable Health-aware Food Recommendation System (MOPI-HFRS). This system generates food recommendations by jointly optimizing three objectives: user preference, personalized healthiness, and nutritional diversity. It also incorporates a reasoning module enhanced by large language models (LLMs) to provide interpretable recommendations that promote healthy dietary knowledge. The framework integrates descriptive features and health data using two structure learning and pooling modules within a graph learning framework. Pareto optimization is applied to balance the multi-faceted objectives. To further enhance healthy dietary knowledge, the system leverages LLMs by infusing knowledge from the recommendation model, generating meaningful interpretations for the recommendations. Extensive experiments on the proposed benchmarks demonstrate that MOPI-HFRS outperforms state-of-the-art methods by delivering diverse, healthy food recommendations alongside reliable explanations.

Subgraph matching is one of the fundamental problems in graph analytics. Existing methods generate matching orders to guide their search, which consists of a series of extensions. Each time, they extend smaller partial matches into larger ones until all complete answers are obtained. However, these methods have two significant drawbacks. Firstly, their matching order generations are usually heuristic and challenging to be effective for different queries. Secondly, each extension, serving as its computation unit, is coarse-grained and may hinder performance. This granularity issue stems from merging generation and expansion operations into a single computation unit. To address these challenges, we introduce a pioneering framework for Reinforced Subgraph Matching (RSM) that features a fine-grained operation-based search plan. Initially, RSM proposes a fresh paradigm for search, referred to as operation-level search, where each computation unit is defined as an operation that either generates or expands a candidate set under a query vertex. To deal with the second problem and fully exploit the potential of this novel search paradigm, RSM implements a reinforcement learning strategy to generate operation-level search plans. RSM's reinforcement learning approach for constructing operation-based search plans encompasses three modules. In the first module, we employ graph neural networks to extract query vertex representation from graphs. Then, the other two modules leverage multilayer perceptron and are designed to create the generation and expansion operations, respectively. Extensive experiments on real-world graph datasets validate that RSM cuts down query processing time, outperforming existing algorithms by up to 1 to 2 orders of magnitude.

Graph Neural Networks (GNNs) have demonstrated their effectiveness in various graph learning tasks, yet their reliance on neighborhood aggregation during inference poses challenges for deployment in latency-sensitive applications, such as real-time financial fraud detection. To address this limitation, recent studies have proposed distilling knowledge from teacher GNNs into student Multi-Layer Perceptrons (MLPs) trained on node content, aiming to accelerate inference. However, these approaches often inadequately explore structural information when inferring unseen nodes. To this end, we introduce SimMLP, a Self-supervised framework for learning MLPs on graphs, designed to fully integrate rich structural information into MLPs. Notably, SimMLP is the first MLP-learning method that can achieve equivalence to GNNs in the optimal case. The key idea is to employ self-supervised learning to align the representations encoded by graph context-aware GNNs and neighborhood dependency-free MLPs, thereby fully integrating the structural information into MLPs. We provide a comprehensive theoretical analysis, demonstrating the equivalence between SimMLP and GNNs based on mutual information and inductive bias, highlighting SimMLP's advanced structural learning capabilities. Additionally, we conduct extensive experiments on 20 benchmark datasets, covering node classification, link prediction, and graph classification, to showcase SimMLP's superiority over state-of-the-art baselines, particularly in scenarios involving unseen nodes (e.g., inductive and cold-start node classification) where structural insights are crucial. Our codes are available at: https://github.com/Zehong-Wang/SimMLP.

Collaborative filtering (CF) is a prevalent technique utilized in recommender systems (RSs), and has been extensively deployed in various real-world applications. A recent study in CF focuses on improving the quality of representations from the perspective of alignment and uniformity on the hyperspheres for enhanced recommendation performance. It promotes alignment to increase the similarity between representations of interacting users and items, and enhances uniformity to have more uniformly distributed user and item representations within their respective hyperspheres. However, although alignment and uniformity are enforced by two different optimized objectives, respectively, they jointly constitute the supervised signals for model training. Models trained with only supervised signals in labeled data can inevitably overfit the noise introduced by label sampling variance, even with i.i.d. datasets. This overfitting to noise further compromises the model's generalizability and performance on unseen testing data. To address this issue, in this study, we aim to mitigate the effect caused by the sampling variance in labeled training data to improve representation generalizability from the perspective of alignment and uniformity. Representations with more generalized alignment and uniformity further lead to improved model performance on testing data. Specifically, we model the data as a user-item interaction bipartite graph, and apply a graph neural network (GNN) to learn the user and item representations. This graph modeling approach allows us to integrate self-supervised signals into the RS, by performing self-supervised contrastive learning on the user and item representations from the perspective of label-irrelevant alignment and uniformity. Since the representations are less dependent on label supervision, they can capture more label-irrelevant data structures and patterns, leading to more generalized alignment and uniformity. We conduct extensive experiments on three benchmark datasets to demonstrate the superiority of our framework (i.e., improved performance and faster convergence speed). Our codes: https://github.com/zyouyang/AUPlus

The success of GNNs highly relies on the accurate labeling of data. Existing methods of ensuring accurate labels, such as weakly-supervised learning, mainly focus on the existing nodes in the graphs. However, in reality, new nodes always continuously emerge on dynamic graphs, with different categories and even label noises. To this end, we formulate a new problem, Graph Cross-Supervised Learning, or Graph Weak-Shot Learning, that describes the challenges of modeling new nodes with novel classes and potential label noises. To solve this problem, we propose Lipshitz-regularized Mixture-of-Experts similarity network (LIME), a novel framework to encode new nodes and handle label noises. Specifically, we first design a node similarity network to capture the knowledge from the original classes, aiming to obtain insights for the emerging novel classes. Then, to enhance the similarity network's generalization to new nodes that could have a distribution shift, we employ the Mixture-of-Experts technique to increase the generalization of knowledge learned by the similarity network. To further avoid losing generalization ability during training, we introduce the Lipschitz bound to stabilize model output and alleviate the distribution shift issue. Empirical experiments validate LIME's effectiveness: we observe a substantial enhancement of up to 11.34% in node classification accuracy compared to the backbone model when subjected to the challenges of label noise on novel classes across five benchmark datasets. The code can be accessed through https://github.com/xiangchi-yuan/Graph-Cross-Supervised-Learning.

The opioid crisis has been one of the most critical society concerns in the United States. Although the medication assisted treatment (MAT) is recognized as the most effective treatment for opioid misuse and addiction, the various side effects can trigger opioid relapse. In addition to MAT, the dietary nutrition intervention has been demonstrated its importance in opioid misuse prevention and recovery. However, research on the alarming connections between dietary patterns and opioid misuse remain under-explored. In response to this gap, in this paper, we first establish a large-scale multifaceted dietary benchmark dataset related to opioid users at the first attempt and then develop a novel framework - i.e., namely Opioid Misuse Detection with INterpretable Dietary Patterns (Diet-ODIN) - to bridge heterogeneous graph (HG) and large language model (LLM) for the identification of users with opioid misuse and the interpretation of their associated dietary patterns. Specifically, in Diet-ODIN, we first construct an HG to comprehensively incorporate both dietary and health-related information, and then we devise a holistic graph learning framework with noise reduction to fully capitalize both users' individual dietary habits and shared dietary patterns for the detection of users with opioid misuse. To further delve into the intricate correlations between dietary patterns and opioid misuse, we exploit an LLM by utilizing the knowledge obtained from the graph learning model for interpretation. The extensive experimental results based on our established benchmark with quantitative and qualitative measures demonstrate the outstanding performance of Diet-ODIN on exploring the complex interplay between opioid misuse and dietary patterns, by comparison with state-of-the-art baseline methods. Our code, built benchmark and system demo are available at https://github.com/JasonZhangzy1757/Diet-ODIN.

Modern machine learning techniques, particularly deep learning, have showcased remarkable efficacy across numerous knowledge discovery and data mining applications. However, the advancement of many of these methods is frequently impeded by resource constraint challenges in many scenarios, such as limited labeled data (data-level), small model size requirements in real-world computing platforms (model-level), and efficient mapping of the computations to heterogeneous target hardware (system-level). Addressing all these factors is crucial for effectively and efficiently deploying developed models across a broad spectrum of real-world systems, including large-scale social network analysis, recommendation systems, and real-time anomaly detection. Therefore, there is a critical need to develop efficient learning techniques to address the challenges posed by resource limitations, whether from data, model/algorithm, or system/hardware perspectives. The proposed international workshop on "Resource-Efficient Learning for Knowledge Discovery (RelKD 2024)" will provide a great venue for academic researchers and industrial practitioners to share challenges, solutions, and future opportunities of resource-efficient learning.

Graph Neural Networks (GNNs) have been a prominent research area and have been widely deployed in various high-stakes applications in recent years, leading to a growing demand for explanations. While existing explainer methods focus on explaining homogeneous and static GNNs, none of them have attempted to explain heterogeneous temporal GNNs. However, in practice, many real-world databases should be represented as heterogeneous temporal graphs (HTGs), which serve as the fundamental data structure for GNN backbone models in applications. To address this gap, in this paper, we propose HTGExplainer, a novel method for explaining heterogeneous temporal GNNs by considering temporal dependencies and preserving heterogeneity when generating subgraphs as explanations. HTGExplainer employs a deep neural network to re-parameterize the generation process of explanations and incorporates effective heterogeneous and temporal edge embeddings to capture informative semantics used for generating explanatory subgraphs. Extensive experiments are conducted on multiple HTG datasets constructed from real-world scenarios, and the results demonstrate the superior performance of HTGExplainer compared to state-of-the-art baselines.

Drug-drug interaction (DDI), as a possible result of drug combination treatment, could lead to adverse physiological reactions and increasing mortality rates of patients. Therefore, predicting potential DDI has always been an important and challenging issue in medical health applications. Owing to the extensive pharmacological research, we can get access to various drug-related features for DDI predictions; however, most of the existing works on DDI prediction do not incorporate comprehensive features to analyze the DDI patterns. Despite the high performance that the existing works have achieved, the incomplete and noisy information generated from limited sources usually leads to sub-optimal performance and poor generalization ability on the unknown DDI pairs. In this work, we propose a holistic framework, namely Multi-modality Feature Optimal Fusion for Drug-Drug Interaction Prediction (MOF-DDI), that incorporates the features from multiple data sources to resolve the DDI predictions. Specifically, the proposed model jointly considers DDIs literature descriptions, biomedical knowledge graphs, and drug molecular structures to make the prediction. To overcome the issue induced by directly aggregating features in different modalities, we bring a new insight by mapping the representations learned from different sources to a unified hidden space before the combination. The empirical results show that MOF-DDI achieves a large performance gain on different DDI datasets compared with multiple state-of-the-art baselines, especially under the inductive setting.

In recent decades, due to the lucrative profits, the crime of drug trafficking has evolved with modern technologies. Social media, as one of the popular online platforms, have become direct-to-consumer intermediaries for illicit drug trafficking communities to promote and trade drugs. These group-wise drug trafficking activities pose significant challenges to public health and safety, requiring urgent measures to address this issue. However, existing works against the imminent problem still face limitations, such as primarily analyzing individual roles from a single perspective, ignoring the group-wise relationships, and requiring sufficient labeled samples for model training. To this end, we propose a novel HyperGraph Contrastive Learning framework called HyGCL-DC that employs hypergraph to model the higher-order relationships among users to detect Drug trafficking Communities. Firstly, we build a hypergraph called Twitter-HyDrug including online user nodes and four types of hyperedges to depict the rich group-wise relationships among these users. Then, we leverage hypergraph neural networks to model the rich relationships among nodes and hyperedges in the drug trafficking hypergraph. Furthermore, we design a hypergraph self-supervised contrast module, which integrates the augmentation from the structure view and the attribute view to enhance hypergraph representation learning over unlabeled data. Finally, we design an end-to-end framework that combines the self-supervised contrastive module and the supervised module to classify online drug trafficking communities. To comprehensively study the online drug trafficking problem and evaluate our model, we conduct extensive experiments over Twitter-HyDrug and three citation benchmark hypergraph datasets to demonstrate the effectiveness of our model. Our new data and source code are available at https://github.com/HyGCL-DC.

Due to the growing importance of using graph neural networks in high-stakes applications, there is a pressing need to interpret the predicted results of these models. Existing methods for explanation have mainly focused on generating sub-graphs comprising important edges for a specific prediction. However, these methods face two issues. Firstly, they lack counterfactual validity as removing the subgraph may not affect the prediction, and generating plausible counterfactual examples has not been adequately explored. Secondly, they cannot be extended to heterogeneous graphs as the complex information involved in such graphs increases the difficulty of generating interpretations. This paper proposes a novel counterfactual learning method, named CF-HGExplainer, for heterogeneous graphs. The method incorporates a semantic-aware attentive pooling strategy for the heterogeneous graph classifier and designs a heterogeneous decision boundaries extraction module to find the common logic for similar graphs based on the extracted graph embeddings from the classifier. Additionally, we propose to greedily perturb nodes and edges based on the distribution of node features and edge plausibility to train a neural network for heterogeneous edge weight learning. Extensive experiments on two public academic datasets demonstrate the effectiveness of CF-HGExplainer compared to state-of-the-art methods on the graph classification task and graph interpretation task.

Modern machine learning techniques, especially deep neural networks, have demonstrated excellent performance for various knowledge discovery and data mining applications. However, the development of many of these techniques still encounters resource constraint challenges in many scenarios, such as limited labeled data (data-level), small model size requirements in real-world computing platforms (model-level), and efficient mapping of the computations to heterogeneous target hardware (system-level). Addressing all of these metrics is critical for the effective and efficient usage of the developed models in a wide variety of real systems, such as large-scale social network analysis, large-scale recommendation systems, and real-time anomaly detection. Therefore, it is desirable to develop efficient learning techniques to tackle challenges of resource limitations from data, model/algorithm, or (and) system/hardware perspectives. The proposed international workshop on "Resource-Efficient Learning for Knowledge Discovery (RelKD 2023)" will provide a great venue for academic researchers and industrial practitioners to share challenges, solutions, and future opportunities of resource-efficient learning.

Graph structure learning (GSL), which aims to learn the adjacency matrix for graph neural networks (GNNs), has shown great potential in boosting the performance of GNNs. Most existing GSL works apply a joint learning framework where the estimated adjacency matrix and GNN parameters are optimized for downstream tasks. However, as GSL is essentially a link prediction task, whose goal may largely differ from the goal of the downstream task. The inconsistency of these two goals limits the GSL methods to learn the potential optimal graph structure. Moreover, the joint learning framework suffers from scalability issues in terms of time and space during the process of estimation and optimization of the adjacency matrix. To mitigate these issues, we propose a graph structure refinement (GSR) framework with a pretrain-finetune pipeline. Specifically, The pre-training phase aims to comprehensively estimate the underlying graph structure by a multi-view contrastive learning framework with both intra- and inter-view link prediction tasks. Then, the graph structure is refined by adding and removing edges according to the edge probabilities estimated by the pre-trained model. Finally, the fine-tuning GNN is initialized by the pre-trained model and optimized toward downstream tasks. With the refined graph structure remaining static in the fine-tuning space, GSR avoids estimating and optimizing graph structure in the fine-tuning phase which enjoys great scalability and efficiency. Moreover, the fine-tuning GNN is boosted by both migrating knowledge and refining graphs. Extensive experiments are conducted to evaluate the effectiveness (best performance on six benchmark datasets), efficiency, and scalability (13.8 times faster using 32.8% GPU memory compared to the best GSL baseline on Cora) of the proposed model.

Graph Neural Networks (GNNs) have demonstrated a great representation learning capability on graph data and have been utilized in various downstream applications. However, real-world data in web-based applications (e.g., recommendation and advertising) always contains bias, preventing GNNs from learning fair representations. Although many works were proposed to address the fairness issue, they suffer from the significant problem of insufficient learnable knowledge with limited attributes after debiasing. To address this problem, we develop Graph-Fairness Mixture of Experts (G-Fame), a novel plug-and-play method to assist any GNNs to learn distinguishable representations with unbiased attributes. Furthermore, based on G-Fame, we propose G-Fame++, which introduces three novel strategies to improve the representation fairness from node representations, model layer, and parameter redundancy perspectives. In particular, we first present the embedding diversified method to learn distinguishable node representations. Second, we design the layer diversified strategy to maximize the output difference of distinct model layers. Third, we introduce the expert diversified method to minimize expert parameter similarities to learn diverse and complementary representations. Extensive experiments demonstrate the superiority of G-Fame and G-Fame++ in both accuracy and fairness, compared to state-of-the-art methods across multiple graph datasets.

The online emergence of multi-modal sharing platforms (e.g., TikTok, Youtube) is powering personalized recommender systems to incorporate various modalities (e.g., visual, textual and acoustic) into the latent user representations. While existing works on multi-modal recommendation exploit multimedia content features in enhancing item embeddings, their model representation capability is limited by heavy label reliance and weak robustness on sparse user behavior data. Inspired by the recent progress of self-supervised learning in alleviating label scarcity issue, we explore deriving self-supervision signals with effectively learning of modality-aware user preference and cross-modal dependencies. To this end, we propose a new Multi-Modal Self-Supervised Learning (MMSSL) method which tackles two key challenges. Specifically, to characterize the inter-dependency between the user-item collaborative view and item multi-modal semantic view, we design a modality-aware interactive structure learning paradigm via adversarial perturbations for data augmentation. In addition, to capture the effects that user’s modality-aware interaction pattern would interweave with each other, a cross-modal contrastive learning approach is introduced to jointly preserve the inter-modal semantic commonality and user preference diversity. Experiments on real-world datasets verify the superiority of our method in offering great potential for multimedia recommendation over various state-of-the-art baselines. The implementation is released at: https://github.com/HKUDS/MMSSL.

Pairwise ranking models have been widely used to address recommendation problems. The basic idea is to learn the rank of users' preferred items through separating items into positive samples if user-item interactions exist, and negative samples otherwise. Due to the limited number of observable interactions, pairwise ranking models face serious class-imbalance issues. Our theoretical analysis shows that current sampling-based methods cause the vertex-level imbalance problem, which makes the norm of learned item embeddings towards infinite after a certain training iterations, and consequently results in vanishing gradient and affects the model inference results. We thus propose an efficient Vital Negative Sampler (VINS) to alleviate the class-imbalance issue for pairwise ranking model, in particular for deep learning models optimized by gradient methods. The core of VINS is a bias sampler with reject probability that will tend to accept a negative candidate with a larger degree weight than the given positive item. Evaluation results on several real datasets demonstrate that the proposed sampling method speeds up the training procedure 30% to 50% for ranking models ranging from shallow to deep, while maintaining and even improving the quality of ranking results in top-N item recommendations.

GitHub, as the largest social coding platform, has attracted an increasing number of cybercriminals to disseminate malware by posting malicious code repositories. To address the imminent problem, some tools were developed to detect malicious repositories based on the code content. However, most of them ignore the rich relational information among repositories and usually require abundant labeled data to train the model. To this end, one effective way is to exploit unlabeled data to pre-train a model which considers both structural relation and code content of repositories, and further transfer the pre-trained model to the downstream tasks with labeled repository data. In this paper, we propose a novel model adversarial contrastive learning on heterogeneous graph (CLA-HG) to detect malicious repository in GitHub. First of all, CLA-HG builds a heterogeneous graph (HG) to comprehensively model repository data. Afterwards, to exploit unlabeled information in HG, CLA-HG introduces a dual-stream graph contrastive learning mechanism that distinguishes both adversarial subgraph pairs and standard subgraph pairs to pre-train graph neural networks using unlabeled data. Finally, the pre-trained model is fine-tuned to the downstream malicious repository detection task enhanced by a knowledge distillation (KD) module. Extensive experiments on two collected datasets from GitHub demonstrate the effectiveness of CLA-HG in comparison with state-of-the-art methods and popular commercial anti-malware products.

Images are commonly used for various information and knowledge applications, such as advertising and recommendation. Automating image caption generation will significantly improve image accessibility. This cross-modal task, which takes image as input and text as output, however, is difficult for learning. Though prior methods achieve good performance for image caption generation, they rely on either supervised learning which requires sufficient labeled data or unsupervised learning which needs external dataset as language pivot. In this paper, we propose SGCL, a novel Scene Graph Contrastive Learning model for self-supervised image caption generation. SGCL adopts the pre-training and fine-tuning pipeline. Specifically, we first apply scene graph generation and objection detection method to encode scene graph and visual information in the image as feature representation. Later, a decoder network based on graph attention network and recurrent neural network is further designed to generate sequential text as caption. To enable contrastive learning in SGCL, we design scene graph augmentations as contrastive views of images and train the model effectively without ground-truth labels through contrastive learning. Additionally, we introduce the pre-trained word embedding and the context projector to enrich the text representation in the decoder network, which benefits model pre-training. Once the pre-training phase is finished, we further fine-tune the model for the image caption generation task with limited labeled data. Extensive experiments on benchmark dataset demonstrate that SGCL outperforms state-of-the-art models (both supervised and unsupervised).

The development of social media (e.g., Twitter) allows users to make speeches with low cost and broad influence. Thus, social media has become a perfect place for users’ malicious behaviors like committing hate crimes, spreading toxic information, abetting crimes, etc. Malicious behaviors are covert and widespread, with potential relevance regarding topic, person, place, and so on. Therefore, it is necessary to develop novel techniques to detect and disrupt malicious behavior on social media effectively. Previous research has shown promising results in extracting semantic text (speech) representation using natural language processing methods. Yet the latent relation between speeches and the connection between users behind speeches is rarely explored. In light of this, we propose a holistic model named Graph adaption BERT (GraphBERT) to detect malicious behaviors on Twitter with both semantic and relational information. Specifically, we first present a novel and a large-scale corpus of tweet data to benefit both graph-based and language-based malicious behavior detection research. Then, we design a novel model GraphBERT to learn comprehensive tweet and user representation with the integration of both semantic information encoded by transformers (i.e., BERT) and relational information encoded by graph neural network. GraphBERT further leverages a weight adaption BERT module implemented between transformer layers to refine tweet embedding using relational information for malicious tweet classification. Finally, the adapted tweet embedding is used with the initial tweet representation to generate user embedding for malicious user detection. The extensive experiments on the collected Twitter data show that our model outperforms the state-of-the-art baseline methods for both tasks (i.e., malicious tweet classification and malicious user detection).

To model graph-structured data, graph learning, in particular deep graph learning with graph neural networks, has drawn much attention in both academic and industrial communities lately. The effectiveness of prevailing graph learning methods usually rely on abundant labeled data for model training. However, it is common that graphs are scarcely labeled since data annotation and labeling on graphs is always time and resource-consuming. Therefore, it is imperative to investigate graph learning with minimal human supervision for the low-resource settings where limited or even no labeled data is available. In this tutorial, we will focus on the state-of-the-art techniques of Graph Minimally-Supervised Learning, in particular a series of weakly-supervised learning, few-shot learning, and self-supervised learning methods on graph-structured data as well as their real-world applications. The objectives of this tutorial are to: (1) formally categorize the problems in graph minimally-supervised learning and discuss the challenges under different learning scenarios; (2) comprehensively review the existing and recent advances of graph minimally-supervised learning; and (3) elucidate open questions and future research directions. This tutorial introduces major topics within minimally-supervised learning and offers a guide to a new frontier of graph learning.

Driven by the exponential increase of software and the advent of the pull-based development system Git, a large amount of open-source software has emerged on various social coding platforms. GitHub, as the largest platform, not only attracts developers and researchers to contribute legitimate software and research-related source code but has also become a popular platform for an increasing number of cybercriminals to perform continuous cyberattacks. Hence, some tools have been developed to learn representations of repositories on GitHub for various related applications (e.g., malicious repository detection) recently. However, most of them merely focus on code content while ignoring the rich relational data among repositories. In addition, they usually require a mass of resources to obtain sufficient labeled data for model training while ignoring the usefully handy unlabeled data. To this end, we propose a novel model Rep2Vec which integrates the code content, the structural relations, and the unlabeled data to learn the repository representations. First, to comprehensively model the repository data, we build a repository heterogeneous graph (Rep-HG) which is encoded by a graph neural network. Afterwards, to fully exploit unlabeled data in Rep-HG, we introduce adversarial attacks to generate more challenging contrastive pairs for the contrastive learning module to train the encoder in node view and meta-path view simultaneously. To alleviate the workload of the encoder against attacks, we further design a dual-stream contrastive learning module that integrates contrastive learning on adversarial graph and original graph together. Finally, the pre-trained encoder is fine-tuned to the downstream task, and further enhanced by a knowledge distillation module. Extensive experiments on the collected dataset from GitHub demonstrate the effectiveness of Rep2Vec in comparison with state-of-the-art methods for multiple repository tasks.

Node classification is of great importance among various graph mining tasks. In practice, real-world graphs generally follow the long-tail distribution, where a large number of classes only consist of limited labeled nodes. Although Graph Neural Networks (GNNs) have achieved significant improvements in node classification, their performance decreases substantially in such a few-shot scenario. The main reason can be attributed to the vast generalization gap between meta-training and meta-test due to the task variance caused by different node/class distributions in meta-tasks (i.e., node-level and class-level variance). Therefore, to effectively alleviate the impact of task variance, we propose a task-adaptive node classification framework under the few-shot learning setting. Specifically, we first accumulate meta-knowledge across classes with abundant labeled nodes. Then we transfer such knowledge to the classes with limited labeled nodes via our proposed task-adaptive modules. In particular, to accommodate the different node/class distributions among meta-tasks, we propose three essential modules to perform node-level, class-level, and task-level adaptations in each meta-task, respectively. In this way, our framework can conduct adaptations to different meta-tasks and thus advance the model generalization performance on meta-test tasks. Extensive experiments on four prevalent node classification datasets demonstrate the superiority of our framework over the state-of-the-art baselines. Our code is provided at https://github.com/SongW-SW/TENT https://github.com/SongW-SW/TENT.

Opioids (e.g., oxycodone and morphine) are highly addictive prescription (aka Rx) drugs which can be easily overprescribed and lead to opioid overdose. Recently, the opioid epidemic is increasingly serious across the US as its related deaths have risen at alarming rates. To combat the deadly opioid epidemic, a state-run prescription drug monitoring program (PDMP) has been established to alleviate the drug over-prescribing problem in the US. Although PDMP provides a detailed prescription history related to opioids, it is still not enough to prevent opioid overdose because it cannot predict over-prescribing risk. In addition, existing machine learning-based methods mainly focus on drug doses while ignoring other prescribing patterns behind patients' historical records, thus resulting in suboptimal performance. To this end, we propose a novel model DDHGNN - Disentangled Dynamic Heterogeneous Graph Neural Network, for over-prescribing prediction. Specifically, we abstract the PDMP data into a dynamic heterogeneous graph which comprehensively depicts the prescribing and dispensing (P&D) relationships. Then, we design a dynamic heterogeneous graph neural network to learn patients' representations. Furthermore, we devise an adversarial disentangler to learn a disentangled representation which is particularly related to the prescribing patterns. Extensive experiments on a 1-year anonymous PDMP data demonstrate that DDHGNN outperforms state-of-the-art methods, revealing its promising future in preventing opioid overdose.

Graph Neural Networks (GNNs) have been shown as promising solutions for collaborative filtering (CF) with the modeling of user-item interaction graphs. The key idea of existing GNN-based recommender systems is to recursively perform the message passing along the user-item interaction edge for refining the encoded embeddings. Despite their effectiveness, however, most of the current recommendation models rely on sufficient and high-quality training data, such that the learned representations can well capture accurate user preference. User behavior data in many practical recommendation scenarios is often noisy and exhibits skewed distribution, which may result in suboptimal representation performance in GNN-based models. In this paper, we propose SHT, a novel Self-Supervised Hypergraph Transformer framework (SHT) which augments user representations by exploring the global collaborative relationships in an explicit way. Specifically, we first empower the graph neural CF paradigm to maintain global collaborative effects among users and items with a hypergraph transformer network. With the distilled global context, a cross-view generative self-supervised learning component is proposed for data augmentation over the user-item interaction graph, so as to enhance the robustness of recommender systems. Extensive experiments demonstrate that SHT can significantly improve the performance over various state-of-the-art baselines. Further ablation studies show the superior representation ability of our SHT recommendation framework in alleviating the data sparsity and noise issues. The source code and evaluation datasets are available at: https://github.com/akaxlh/SHT.

Graph few-shot learning seeks to alleviate the label scarcity problem resulting from the difficulties and high cost of data annotations in graph learning. However, the overwhelming solutions in graph few-shot learning focus on homogeneous graphs, ignoring the ubiquitous heterogeneous graphs (HGs), which represent real-world complex systems and domain knowledge with multi-typed nodes interconnected by multi-typed edges. To this end, we study the cross-domain few-shot learning problem over HGs and develop a novel model for Cross-domain Heterogeneous Graph Meta learning (CrossHG-Meta). The general idea is to promote the HG node classification in the data-scarce target domain by transferring meta-knowledge from a series of HGs in data-rich source domains. The key challenges are to 1) combat the heterogeneity in HGs to acquire the transferable meta-knowledge; 2) handle the domain shifts between the source HG and target HG; and 3) fast adapt to novel target tasks with few-shot annotated examples. Regarding the graph heterogeneity, CrossHG-Meta firstly builds a graph encoder to aggregate heterogeneous neighborhood information from multiple semantic contexts. Secondly, to tackle domain shifts, a cross-domain meta-learning strategy is proposed to include a domain critic, which is designed to explicitly lead cross-domain adaptation for meta-tasks in different domains and improve model generalizability. Last, to further alleviate data scarcity, CrossHG-Meta leverages unlabelled information in source domains with auxiliary self-supervised learning task to provide cross-domain contrastive regularization alongside the meta-optimization process to facilitate node embedding. Extensive experimental results on three multi-domain HG datasets demonstrate that the proposed model outperforms various state-of-the-art baselines for multiple few-shot node classification tasks under the cross-domain setting.

Knowledge graphs (KGs) are serving as important resources for many applications, such as semantic search, question answering, or dialogue generation. As one of the fundamental tasks, multi-hop KG reasoning aims at generating effective and explainable relation prediction through reasoning paths. The current methods often require sufficient amount of training data (i.e., fact triples) for each query relation, impairing their applicabilities and performances over few-shot relations (with limited data) which are common in KGs. Despite that some few-shot relation reasoning methods have been proposed, their effectiveness and efficiency remain to be improved. To address these challenges, we propose a novel model called ADK-KG for multi-hop few-shot relation reasoning over KGs. In ADK-KG, we introduce a reinforcement learning framework to model the sequential reasoning process. We further develop a text-enhanced heterogeneous graph neural network to encode node embeddings, where entity and relation embeddings are pre-trained using content information. Later, we employ a task-aware meta-learning algorithm to optimize the model parameters that could be fast adapted for few-shot relations. A knowledge distillation module is further designed to make use of unlabeled data for improving model training. Extensive experiments on three benchmark datasets demonstrate that ADK-KG has satisfactory efficiency and outperforms state-of-the-art approaches.

Graph neural networks (GNNs) have been broadly studied on dynamic graphs for their representation learning, majority of which focus on graphs with homogeneous structures in the spatial domain. However, many real-world graphs - i.e., heterogeneous temporal graphs (HTGs) - evolve dynamically in the context of heterogeneous graph structures. The dynamics associated with heterogeneity have posed new challenges for HTG representation learning. To solve this problem, in this paper, we propose heterogeneous temporal graph neural network (HTGNN) to integrate both spatial and temporal dependencies while preserving the heterogeneity to learn node representations over HTGs. Specifically, in each layer of HTGNN, we propose a hierarchical aggregation mechanism, including intra-relation, inter-relation, and across-time aggregations, to jointly model heterogeneous spatial dependencies and temporal dimensions. To retain the heterogeneity, intra-relation aggregation is first performed over each slice of HTG to attentively aggregate information of neighbors with the same type of relation, and then intra-relation aggregation is exploited to gather information over different types of relations; to handle temporal dependencies, across-time aggregation is conducted to exchange information across different graph slices over the HTG. The proposed HTGNN is a holistic framework tailored heterogeneity with evolution in time and space for HTG representation learning. Extensive experiments are conducted on the HTGs built from different real-world datasets and promising results demonstrate the outstanding performance of HTGNN by comparison with state-of-the-art baselines. Our built HTGs and code have been made publicly accessible at: https://github.com/YesLab-Code/HTGNN.

Prevailing supervised graph neural networks suffer from potential performance degradation in the label sparsity case. Though increasing attention has been paid to graph few-shot learning methods for learning effective graph embeddings under the scarcity of labeled data, most existing works study homogeneous graphs while ignoring the ubiquitousness of heterogeneous graphs (HG), where multi-typed nodes are interconnected by multi-typed edges. To this end, we propose to tackle few-shot learning on HG and develop a novel model for Heterogeneous Graph Meta-learning (a.k.a. HG-Meta). Regarding the graph heterogeneity, HG-Meta firstly builds a graph encoder to aggregate heterogeneous neighbors information from multiple semantic contexts (generated by meta-paths). Secondly, to train the graph encoder with meta-learning in a few-shot scenario, HG-Meta tackles meta-task differences produced from meta-task sampling procedure on HG with a task feature scaling module and a degree based task attention module. To further alleviate low-data problem, HG-Meta leverages unlabelled information in HG with auxiliary self-supervised learning task alongside the meta-optimization process to facilitate node embedding. Extensive experiments on two HG datasets demonstrate that HG-Meta outperforms state-of-the-art methods for multiple few-shot node classification tasks.

Relation learning, widely used in recommendation systems or relevant entity search over knowledge graphs, has attracted increasing attentions in recent years. Existing methods like network embedding and graph neural networks (GNNs), learn the node representations from neighbors and calculate the similarity score for relation prediction. Despite effective prediction performance, they lack explanations to the predicted results. We propose a novel interpretable relation learning model named IRL, which can not only predict whether relations exist between node pairs, but also make the inference more transparent and convincing. Specifically, we introduce a meta-path based path encoder to model sequential dependency between nodes through recurrent neural network. We also apply the self-supervised GNN on the extracted sub-graph to capture the graph structure by aggregating information from neighbors, which are fed into the meta-path encoder. In addition, we propose a meta-path walk pruning strategy for positive path generation and an adaptive negative sampling method for negative path generation to improve the quality of paths, which both consider the semantics of nodes in the heterogeneous graph. We conduct extensive experiments on two public heterogeneous graph data, AMiner and Delve, for different relation prediction tasks, which demonstrate significant improvements of our model over the existing embedding-based and sequential modeling-based methods.

People rely on task management applications and digital assistants to capture and track their tasks, and help with executing them. The burden of organizing and scheduling time for tasks continues to reside with users of these systems, despite the high cognitive load associated with these activities. Users stand to benefit greatly from a task management system capable of prioritizing their pending tasks, thus saving them time and effort. In this article, we make three main contributions. First, we propose the problem of task prioritization, formulating it as a ranking over a user’s pending tasks given a history of previous interactions with a task management system. Second, we perform an extensive analysis on the large-scale anonymized, de-identified logs of a popular task management application, deriving a dataset of grounded, real-world tasks from which to learn and evaluate our proposed system. We also identify patterns in how people record tasks as complete, which vary consistently with the nature of the task. Third, we propose a novel contextual deep learning solution capable of performing personalized task prioritization. In a battery of tests, we show that this approach outperforms several operational baselines and other sequential ranking models from previous work. Our findings have implications for understanding the ways people prioritize and manage tasks with digital tools, and in the design of support for users of task management applications.

Learning effective representations for recipes is essential in food studies for recommendation, classification, and other applications. Unlike what has been developed for learning textual or cross-modal embeddings for recipes, the structural relationship among recipes and food items are less explored. In this paper, we formalize the problem recipe representation learning with networks to involve both the textual feature and the structural relational feature into recipe representations. Specifically, we first present RecipeNet, a new and large-scale corpus of recipe data to facilitate network based food studies and recipe representation learning research. We then propose a novel heterogeneous recipe network embedding model, rn2vec, to learn recipe representations. The proposed model is able to capture textual, structural, and nutritional information through several neural network modules, including textual CNN, inner-ingredients transformer, and a graph neural network with hierarchical attention. We further design a combined objective function of node classification and link prediction to jointly optimize the model. The extensive experiments show that our model outperforms state-of-the-art baselines on two classic food study tasks. Dataset and codes are available at https://github.com/meettyj/rn2vec.

Prescription (aka Rx) drugs can be easily overprescribed and lead to drug abuse or opioid overdose. Accordingly, a state-run prescription drug monitoring program (PDMP) in the United States has been developed to reduce Overprescribing. However, PDMP has limited capability in detecting patients' potential overprescribing behaviors, impairing its effectiveness in preventing drug abuse and overdose in patients. Despite a few machine-learning-based methods that have been proposed for detecting overprescribing, they usually ignore the patient prescribing behavior and their performances are not satisfying. In light of this, we propose a novel model RxNet for overprescribing detection in PDMP. RxNet builds a dynamic heterogeneous graph to model Rx refills that are essentially prescribing and dispensing (P&D) relationships among various Rx entries (e.g., patients) whose representations are encoded by graph neural network. In addition, to explore the dynamic Rx-refill behavior and medical condition variation of patients, an RxLSTM network is designed to update representations of patients. Based on the output of RxLSTM, a dosing-adaptive network is leveraged to extract and recalibrate dosing patterns and obtain the refined patient representations which are finally utilized for overprescribing detection. The extensive experimental results on a 1-year Ohio PDMP data demonstrate that RxNet consistently outperforms state-of-the-art methods in predicting patients at high risk of opioid overdose and drug abuse, with an average of 5.7% and 7.3% improvement on F1 score respectively.

Prevailing methods of graph representation learning (GRL) usually rely on learning from "big'' data, requiring a large amount of labeled data for model training. However, it is common that graphs are associated with "small'' labeled data as data annotation and labeling is always a time and resource consuming task. The fact overshadows GRL's capability and applicability for many real situations. Therefore, data efficient learning on graphs has become essential for many real-world applications and there have been many studies working on this topic in recent years. In this tutorial, we will systematically review recent studies of data efficient learning on graphs, in particular a series of methods and applications of graph few-shot learning and graph self-supervised learning. At first, we will introduce the overview of graph representation learning methods, conventional few-shot learning, and self-supervised learning techniques. Then, we will present the work of data efficient learning on graphs in terms of three major graph mining tasks at different granularity levels: node-level learning tasks, graph-level learning tasks, and edge-level learning tasks. In the end, we will conclude the tutorial and raise open problems and pressing issues in future research. The authors of this tutorial are active and productive researchers in this research area.

The recent success of graph neural networks has significantly boosted molecular property prediction, advancing activities such as drug discovery. The existing deep neural network methods usually require large training dataset for each property, impairing their performance in cases (especially for new molecular properties) with a limited amount of experimental data, which are common in real situations. To this end, we propose Meta-MGNN, a novel model for few-shot molecular property prediction. Meta-MGNN applies molecular graph neural network to learn molecular representations and builds a meta-learning framework for model optimization. To exploit unlabeled molecular information and address task heterogeneity of different molecular properties, Meta-MGNN further incorporates molecular structures, attribute based self-supervised modules and self-attentive task weights into the former framework, strengthening the whole learning model. Extensive experiments on two public multi-property datasets demonstrate that Meta-MGNN outperforms a variety of state-of-the-art methods.

Web personalization, e.g., recommendation or relevance search, tailoring a service/product to accommodate specific online users, is becoming increasingly important. Inductive personalization aims to infer the relations between existing entities and unseen new ones, e.g., searching relevant authors for new papers or recommending new items to users. This problem, however, is challenging since most of recent studies focus on transductive problem for existing entities. In addition, despite some inductive learning approaches have been introduced recently, their performance is sub-optimal due to relatively simple and inflexible architectures for aggregating entity’s content. To this end, we propose the inductive contextual personalization (ICP) framework through contextual relation learning. Specifically, we first formulate the pairwise relations between entities with a ranking optimization scheme that employs neural aggregator to fuse entity’s heterogeneous contents. Next, we introduce a node embedding term to capture entity’s contextual relations, as a smoothness constraint over the prior ranking objective. Finally, the gradient descent procedure with adaptive negative sampling is employed to learn the model parameters. The learned model is capable of inferring the relations between existing entities and inductive ones. Thorough experiments demonstrate that ICP outperforms numerous baseline methods for two different applications, i.e., relevant author search and new item recommendation.

With the recent advancement of deep learning, molecular representation learning -- automating the discovery of feature representation of molecular structure, has attracted significant attention from both chemists and machine learning researchers. Deep learning can facilitate a variety of downstream applications, including bio-property prediction, chemical reaction prediction, etc. Despite the fact that current SMILES string or molecular graph molecular representation learning algorithms (via sequence modeling and graph neural networks, respectively) have achieved promising results, there is no work to integrate the capabilities of both approaches in preserving molecular characteristics (e.g, atomic cluster, chemical bond) for further improvement. In this paper, we propose GraSeq, a joint graph and sequence representation learning model for molecular property prediction. Specifically, GraSeq makes a complementary combination of graph neural networks and recurrent neural networks for modeling two types of molecular inputs, respectively. In addition, it is trained by the multitask loss of unsupervised reconstruction and various downstream tasks, using limited size of labeled datasets. In a variety of chemical property prediction tests, we demonstrate that our GraSeq model achieves better performance than state-of-the-art approaches.

In today's information and computational society, complex systems are often modeled as multi-modal networks associated with heterogeneous structural relation, unstructured attribute/content, temporal context, or their combinations. The abundant information in multi-modal network requires both a domain understanding and large exploratory search space when doing feature engineering for building customized intelligent solutions in response to different purposes. Therefore, automating the feature discovery through representation learning in multi-modal networks has become essential for many applications. In this tutorial, we systematically review the area of multi-modal network representation learning, including a series of recent methods and applications. These methods will be categorized and introduced in the perspectives of unsupervised, semi-supervised and supervised learning, with corresponding real applications respectively. In the end, we conclude the tutorial and raise open discussions. The authors of this tutorial are active and productive researchers in this area.

Complex systems in different disciplines are usually modeled as heterogeneous networks. Different from homogeneous networks or attributed networks, heterogeneous networks are associated with complexity in heterogeneous structure or heterogeneous content or both. The abundant information in heterogeneous networks provide opportunities yet pose challenges for researchers and practitioners to develop customized machine learning solutions for solving different problems in complex systems. We are motivated to do significant work for learning from heterogeneous networks. In this paper, we first introduce the motivation and background of this research. Later, we present our current work which include a series of proposed methods and applications. These methods will be introduced in the perspectives of personalization in web-based systems and heterogeneous network embedding. In the end, we raise several research directions as future agenda.

Spatial event forecasting is challenging and crucial for urban sensing scenarios, which is beneficial for a wide spectrum of spatial-temporal mining applications, ranging from traffic management, public safety, to environment policy making. In spite of significant progress has been made to solve spatial-temporal prediction problem, most existing deep learning based methods based on a coarse-grained spatial setting and the success of such methods largely relies on data sufficiency. In many real-world applications, predicting events with a fine-grained spatial resolution do play a critical role to provide high discernibility of spatial-temporal data distributions. However, in such cases, applying existing methods will result in weak performance since they may not well capture the quality spatial-temporal representations when training triple instances are highly imbalanced across locations and time.

Traffic accident forecasting is a vital part of intelligent transportation systems in urban sensing. However, predicting traffic accidents is not trivial because of two key challenges: i) the complexities of external factors which are presented with heterogeneous data structures; ii) the complex sequential transition regularities exhibited with time-dependent and high-order inter-correlations. To address these challenges, we develop a deep Dynamic Fusion Network framework (DFN), to explore the central theme of improving the ability of deep neural network on modeling heterogeneous external factors in a fully dynamic manner for traffic accident forecasting. Specifically, DFN first develops an integrative architecture, i.e., with the cooperation of a context-aware embedding module and a hierarchical fusion network, to effectively transferring knowledge from different external units for spatial-temporal pattern learning across space and time. After that, we further develop a temporal aggregation neural network layer to automatically capture relevance scores from the temporal dimension. Through extensive experiments on real-world data collected from New York City, we validate the effectiveness of our framework against various competitive methods. Besides, we also provide a qualitative analysis on prediction results to show the model interpretability.

Online purchase forecasting is of great importance in e-commerce platforms, which is the basis of how to present personalized interesting product lists to individual customers. However, predicting online purchases is not trivial as it is influenced by many factors including: (i) the complex temporal pattern with hierarchical inter-correlations; (ii) arbitrary category dependencies. To address these factors, we develop a Graph Multi-Scale Pyramid Networks (GMP) framework to fully exploit users' latent behavioral patterns with both multi-scale temporal dynamics and arbitrary inter-dependencies among product categories. In GMP, we first design a multi-scale pyramid modulation network architecture which seamlessly preserves the underlying hierarchical temporal factors--governing users' purchase behaviors. Then, we employ convolution recurrent neural network to encode the categorical temporal pattern at each scale. After that, we develop a resolution-wise recalibration gating mechanism to automatically re-weight the importance of each scale-view representations. Finally, a context-graph neural network module is proposed to adaptively uncover complex dependencies among category-specific purchases. Extensive experiments on real-world e-commerce datasets demonstrate the superior performance of our method over state-of-the-art baselines across various settings.

Representation learning in heterogeneous graphs aims to pursue a meaningful vector representation for each node so as to facilitate downstream applications such as link prediction, personalized recommendation, node classification, etc. This task, however, is challenging not only because of the demand to incorporate heterogeneous structural (graph) information consisting of multiple types of nodes and edges, but also due to the need for considering heterogeneous attributes or contents (e.g., text or image) associated with each node. Despite a substantial amount of effort has been made to homogeneous (or heterogeneous) graph embedding, attributed graph embedding as well as graph neural networks, few of them can jointly consider heterogeneous structural (graph) information as well as heterogeneous contents information of each node effectively. In this paper, we propose HetGNN, a heterogeneous graph neural network model, to resolve this issue. Specifically, we first introduce a random walk with restart strategy to sample a fixed size of strongly correlated heterogeneous neighbors for each node and group them based upon node types. Next, we design a neural network architecture with two modules to aggregate feature information of those sampled neighboring nodes. The first module encodes "deep" feature interactions of heterogeneous contents and generates content embedding for each node. The second module aggregates content (attribute) embeddings of different neighboring groups (types) and further combines them by considering the impacts of different groups to obtain the ultimate node embedding. Finally, we leverage a graph context loss and a mini-batch gradient descent procedure to train the model in an end-to-end manner. Extensive experiments on several datasets demonstrate that HetGNN can outperform state-of-the-art baselines in various graph mining tasks, i.e., link prediction, recommendation, node classification & clustering and inductive node classification & clustering.

Representation learning in heterogeneous networks faces challenges due to heterogeneous structural information of multiple types of nodes and relations, and also due to the unstructured attribute or content (e.g., text) associated with some types of nodes. While many recent works have studied homogeneous, heterogeneous, and attributed networks embedding, there are few works that have collectively solved these challenges in heterogeneous networks. In this paper, we address them by developing a Semantic-aware Heterogeneous Network Embedding model (SHNE). SHNE performs joint optimization of heterogeneous SkipGram and deep semantic encoding for capturing both heterogeneous structural closeness and unstructured semantic relations among all nodes, as function of node content, that exist in the network. Extensive experiments demonstrate that SHNE outperforms state-of-the-art baselines in various heterogeneous network mining tasks, such as link prediction, document retrieval, node recommendation, relevance search, and class visualization.

Citywide abnormal events, such as crimes and accidents, may result in loss of lives or properties if not handled efficiently. It is important for a wide spectrum of applications, ranging from public order maintaining, disaster control and people's activity modeling, if abnormal events can be automatically predicted before they occur. However, forecasting different categories of citywide abnormal events is very challenging as it is affected by many complex factors from different views: (i) dynamic intra-region temporal correlation; (ii) complex inter-region spatial correlations; (iii) latent cross-categorical correlations. In this paper, we develop a Multi-View and Multi-Modal Spatial-Temporal learning (MiST) framework to address the above challenges by promoting the collaboration of different views (spatial, temporal and semantic) and map the multi-modal units into the same latent space. Specifically, MiST can preserve the underlying structural information of multi-view abnormal event data and automatically learn the importance of view-specific representations, with the integration of a multi-modal pattern fusion module and a hierarchical recurrent framework. Extensive experiments on three real-world datasets, i.e., crime data and urban anomaly data, demonstrate the superior performance of our MiST method over the state-of-the-art baselines across various settings.

In this paper, we study the problem of author identification in big scholarly data, which is to effectively rank potential authors for each anonymous paper by using historical data. Most of the existing de-anonymization approaches predict relevance score of paper-author pair via feature engineering, which is not only time and storage consuming, but also introduces irrelevant and redundant features or miss important attributes. Representation learning can automate the feature generation process by learning node embeddings in academic network to infer the correlation of paper-author pair. However, the learned embeddings are often for general purpose (independent of the specific task), or based on network structure only (without considering the node content). To address these issues and make a further progress in solving the author identification problem, we propose Camel, a content-aware and meta-path augmented metric learning model. Specifically, first, the directly correlated paper-author pairs are modeled based on distance metric learning by introducing a push loss function. Next, the paper content embedding encoded by the gated recurrent neural network is integrated into the distance loss. Moreover, the historical bibliographic data of papers is utilized to construct an academic heterogeneous network, wherein a meta-path guided walk integrative learning module based on the task-dependent and content-aware Skipgram model is designed to formulate the correlations between each paper and its indirect author neighbors, and further augments the model. Extensive experiments demonstrate that Camel outperforms the state-of-the-art baselines. It achieves an average improvement of 6.3% over the best baseline method.

Data sparsity and cold-start problems are prevalent in recommender systems. To address such problems, both the observable explicit social information (e.g., user-user trust connections) and the inferable implicit correlations (e.g., implicit neighbors computed by similarity measurement) have been introduced to complement user-item ratings data for improving the performances of traditional model-based recommendation algorithms such as matrix factorization. Although effective, (1) the utilization of the explicit user-user social relationships suffers from the weakness of unavailability in real systems such as Netflix or the issue of sparse observable content like 0.03% trust density in Epinions, thus there is no or little explicit social information that can be employed to improve baseline model in real applications; (2) the current similarity measurement approaches focus on inferring implicit correlations between a user (item) and their direct neighbors or top-k similar neighbors based on user-item ratings bipartite network, so that they fail to comprehensively unfold the indirect potential relationships among users and items. To solve these issues regarding both explicit/implicit social recommendation algorithms, we design a joint model of matrix factorization and implicit walk integrative learning, i.e., ImWalkMF, which only uses explicit ratings information yet models both direct rating feedbacks and multiple direct/indirect implicit correlations among users and items from a random walk perspective. We further propose a combined strategy for training two independent components in the proposed model based on sampling. The experimental results on two real-world sparse datasets demonstrate that ImWalkMF outperforms the traditional regularized/probabilistic matrix factorization models as well as other competitive baselines that utilize explicit/implicit social information.

To address the issue of data sparsity and cold-start in recommender system, social information (e.g., user-user trust links) has been introduced to complement rating data for improving the performances of traditional model-based recommendation techniques such as matrix factorization (MF) and Bayesian personalized ranking (BPR). Although effective, the utilization of the explicit user-user relationships extracted directly from such social information has three main limitations. First, it is difficult to obtain explicit and reliable social links. Only a small portion of users indicate explicitly their trusted friends in recommender systems. Second, the “cold-start” users are “cold” not only on rating but also on socializing. There is no significant amount of explicit social information that can be useful for “cold-start” users. Third, an active user can be socially connected with others who have different taste/preference. Direct usage of explicit social links may mislead recommendation. To address these issues, we propose to extract implicit and reliable social information from user feedbacks and identify top-k semantic friends for each user. We incorporate the top-k semantic friends information into MF and BPR frameworks to solve the problems of ratings prediction and items ranking, respectively. The experimental results on three real-world datasets show that our proposed approaches achieve better results than the state-of-the-art MF with explicit social links (with 3.0% improvement on RMSE), and social BPR (with 9.1% improvement on AUC).