Zhao Li

TL;DR: This paper proposes AddGraph, a general end-to-end anomalous edge detection framework using an extended temporal GCN (Graph Convolutional Network) with an attention model, which can capture both long- term patterns and the short-term patterns in dynamic graphs.

TL;DR: In this article, a contrastive self-supervised learning framework for anomaly detection on attributed networks is proposed, which exploits the local information from network data by sampling a novel type of contrastive instance pair, which can capture the relationship between each node and its neighboring substructure.

TL;DR: An algorithm called Nonnegative Matrix Factorization on Orthogonal Subspace (NMFOS), in which the generation of orthogonal factor matrices is part of objective function minimization is proposed, and orthogonality is achieved without resorting to additional constraints, and the computational complexity is decreased.

TL;DR: In this article, a feature induced partial multi-label learning (fPML) approach is proposed, which simultaneously estimates noisy labels and trains multilabel classifiers by factorizing the observed instance-label association matrix and the instance-feature matrix into low-rank matrices.

TL;DR: A novel method with Hierarchical bipartite Graph Neural Network (HiGNN) to handle large-scale e-commerce tasks by stacking multiple GNN modules and using a deterministic clustering algorithm alternately, HiGNN is able to efficiently obtain hierarchical user and item embeddings simultaneously, and effectively predict user preferences on a larger scale.