A5100740622- Advanced Graph Neural Networks
- Topic Modeling
- Natural Language Processing Techniques
- Recommender Systems and Techniques
- Complex Network Analysis Techniques
- Anomaly Detection Techniques and Applications
- Anaerobic Digestion and Biogas Production
- Network Security and Intrusion Detection
- Sentiment Analysis and Opinion Mining
- Advanced Text Analysis Techniques
- Semantic Web and Ontologies
- Domain Adaptation and Few-Shot Learning
- Advanced Bandit Algorithms Research
- Biofuel production and bioconversion
- Advanced materials and composites
- Advanced Wireless Communication Techniques
- Privacy-Preserving Technologies in Data
- Text and Document Classification Technologies
- Machine Learning in Materials Science
- Topological and Geometric Data Analysis
- Software Engineering Research
- Adversarial Robustness in Machine Learning
- Software Testing and Debugging Techniques
- Educational Technology and Assessment
- Heat Transfer and Optimization
Beihang University
2013-2025
Northeast Agricultural University
2022-2025
Shanghai Jiao Tong University
2024-2025
Zhejiang Normal University
2024-2025
University of Electronic Science and Technology of China
2023-2024
University of California, Riverside
2020-2024
California Polytechnic State University
2024
California State University, Long Beach
2024
Wuhan Polytechnic University
2018-2024
China National Petroleum Corporation (China)
2022-2024
Recommender systems have become prosperous nowadays, designed to predict users’ potential interests in items by learning embeddings. Recent developments of the Graph Neural Networks (GNNs) also provide recommender (RSs) with powerful backbones learn embeddings from a user-item graph. However, only leveraging interactions suffers cold-start issue due difficulty data collection. Hence, current endeavors propose fusing social information alleviate it, which is recommendation problem. Existing...
Sequential recommendation models the dynamics of a user's previous behaviors in order to forecast next item, and has drawn lot attention. Transformer-based approaches, which embed items as vectors use dot-product self-attention measure relationship between items, demonstrate superior capabilities among existing sequential methods. However, users' real-world are uncertain rather than deterministic, posing significant challenge present techniques. We further suggest that dot-product-based...
Research on social bot detection plays a crucial role in maintaining the order and reliability of information dissemination while increasing trust interactions. The current mainstream models rely black-box neural network technology, e.g., Graph Neural Network, Transformer, etc., which lacks interpretability. In this work, we present UnDBot, novel unsupervised, interpretable, yet effective practical framework for detecting bots. This is built upon structural theory. We begin by designing...
Relation classification is an important research arena in the field of natural language processing (NLP). In this paper, we present SDP-LSTM, a novel neural network to classify relation two entities sentence. Our architecture leverages shortest dependency path (SDP) between entities; multichannel recurrent networks, with long short term memory (LSTM) units, pick up heterogeneous information along SDP. proposed model has several distinct features: (1) The paths retain most relevant (to...
Graph Neural Networks (GNNs) are de facto solutions to structural data learning. However, it is susceptible low-quality and unreliable structure, which has been a norm rather than an exception in real-world graphs. Existing graph structure learning (GSL) frameworks still lack robustness interpretability. This paper proposes general GSL framework, SE-GSL, through entropy the hierarchy abstracted encoding tree. Particularly, we exploit one-dimensional maximize embedded information content when...
Community detection is a critical task in graph theory, social network analysis, and bioinformatics, where communities are defined as clusters of densely interconnected nodes. However, detecting large-scale networks with millions nodes billions edges remains challenging due to the inefficiency unreliability existing methods. Moreover, many current approaches limited specific types, such unweighted or undirected graphs, reducing their broader applicability. To address these issues, we propose...
Neural network language models (NNLMs) have attracted a lot of attention recently. In this paper, we present training method that can incrementally train the hierarchical softmax function for NNMLs. We split cost to model old and update corpora separately, factorize objective softmax. Then provide new stochastic gradient based all word vectors parameters, by comparing tree generated on corpus combined (old update) corpus. Theoretical analysis shows mean square error parameter be bounded...
Learning representations for graphs plays a critical role in wide spectrum of downstream applications. In this paper, we summarize the limitations prior works three folds: representation space, modeling dynamics and uncertainty. To bridge gap, propose to learn dynamic graph hyperbolic first time, which aims infer stochastic node representations. Working with present novel Hyperbolic Variational Graph Neural Network, referred as HVGNN. particular, model dynamics, introduce Temporal GNN (TGNN)...
Graph anomaly detection in this paper aims to distinguish abnormal nodes that behave differently from the benign ones accounting for majority of graph-structured instances. Receiving increasing attention both academia and industry, yet existing research on task still suffers two critical issues when learning informative anomalous behavior graph data. For one thing, anomalies are usually hard capture because their subtle shortage background knowledge about them, which causes severe sample...
Abstract Graph level anomaly detection (GLAD) aims to spot anomalous graphs that structure pattern and feature information are different from most normal in a graph set, which is rarely studied by other researchers but has significant application value. For instance, GLAD can be used distinguish some characteristic molecules drug discovery chemical analysis. However, mainly faces the following three challenges: (1) learning more comprehensive representations differ abnormal graphs, (2)...
Continual graph learning routinely finds its role in a variety of real-world applications where the data with different tasks come sequentially. Despite success prior works, it still faces great challenges. On one hand, existing methods work zero-curvature Euclidean space, and largely ignore fact that curvature varies over com- ing sequence. other continual learners literature rely on abundant labels, but labeling practice is particularly hard especially for continuously emerging graphs...
Topology-imbalance is a graph-specific imbalance problem caused by the uneven topology positions of labeled nodes, which significantly damages performance GNNs. What topology-imbalance means and how to measure its impact on graph learning remain under-explored. In this paper, we provide new understanding from global view supervision information distribution in terms under-reaching over-squashing, motivates two quantitative metrics as measurements. light our analysis, propose novel...