Deep adaptive collaborative graph neural network for social recommendation

Article Computer Science, Artificial Intelligence

Mutually reinforced network embedding: An integrated approach to research paper recommendation

Xin Mei et al.

Summary: This paper discusses the cold start problem in research paper recommendation and proposes a new method that considers network structure, textual information, and co-authorship of papers. A mutual reinforcement network embedding model is developed to enhance the recommendation. Experimental results demonstrate the effectiveness of the proposed method.

EXPERT SYSTEMS WITH APPLICATIONS (2022)

Add to Collection

Article Computer Science, Artificial Intelligence

Deep Collaborative Filtering with Multi-Aspect Information in Heterogeneous Networks

Chuan Shi et al.

Summary: In this paper, a Neural network based Aspect-level Collaborative Filtering model (NeuACF) is proposed to exploit different aspect latent factors. By modeling the rich object properties and relations in recommender system as a heterogeneous information network, NeuACF first extracts different aspect-level similarity matrices of users and items through different meta-paths, then feeds an elaborately designed deep neural network to learn aspect-level latent factors. The aspect-level latent factors are fused for top-N recommendation.

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING (2021)

Add to Collection

Proceedings Paper Computer Science, Artificial Intelligence

Hypergraph Convolutional Network for Group Recommendation

Renqi Jia et al.

Summary: Group activities are becoming increasingly common, leading to a greater need for research in group recommendation techniques. Current methods face limitations in capturing general group preferences independent of individual user interests. The proposed dual channel Hypergraph Convolutional network for group Recommendation addresses these limitations and consistently outperforms state-of-the-art techniques according to experimental results.

2021 21ST IEEE INTERNATIONAL CONFERENCE ON DATA MINING (ICDM 2021) (2021)

Add to Collection

Proceedings Paper Computer Science, Artificial Intelligence

Interest-aware Message-Passing GCN for Recommendation

Fan Liu et al.

Summary: We propose a novel Interest-aware Message-Passing GCN (IMP-GCN) recommendation model, which tackles the over-smoothing problem in recommendation by performing high-order graph convolution inside subgraphs. By designing an unsupervised subgraph generation module, we are able to effectively identify users with common interests and avoid propagating negative information. Experimental results demonstrate that our model achieves performance improvement by stacking more layers.

PROCEEDINGS OF THE WORLD WIDE WEB CONFERENCE 2021 (WWW 2021) (2021)

Add to Collection

Proceedings Paper Computer Science, Artificial Intelligence

Socially-Aware Self-Supervised Tri-Training for Recommendation

Junliang Yu et al.

Summary: The paper introduces a general socially-aware SSL framework with tri-training that can extract self-supervision signals from user social information, improving the performance of recommendation systems.

KDD '21: PROCEEDINGS OF THE 27TH ACM SIGKDD CONFERENCE ON KNOWLEDGE DISCOVERY & DATA MINING (2021)

Add to Collection

Proceedings Paper Computer Science, Artificial Intelligence

Self-Supervised Multi-Channel Hypergraph Convolutional Network for Social Recommendation

Junliang Yu et al.

Summary: This paper proposes a multi-channel hypergraph convolutional network to enhance social recommendation by leveraging high-order user relations, obtaining comprehensive user representations for recommendation results. Additionally, by integrating self-supervised learning and hierarchical mutual information maximization, the model compensates for aggregating losses and regains connectivity information effectively.

PROCEEDINGS OF THE WORLD WIDE WEB CONFERENCE 2021 (WWW 2021) (2021)

Add to Collection

Proceedings Paper Computer Science, Information Systems

Social Recommendation with Implicit Social Influence

Changhao Song et al.

Summary: This work discusses the importance of explicit and implicit influences in social recommendation, improving existing methods by modeling local implicit influence and global implicit influence. The effectiveness of the proposed model is demonstrated in predicting unobserved social relationships and personalizing global popularity.

SIGIR '21 - PROCEEDINGS OF THE 44TH INTERNATIONAL ACM SIGIR CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION RETRIEVAL (2021)

Add to Collection

Proceedings Paper Computer Science, Information Systems

ConsisRec: Enhancing GNN for Social Recommendation via Consistent Neighbor Aggregation

Liangwei Yang et al.

Summary: Social recommendation addresses the cold-start problem in rating prediction by leveraging Graph Neural Networks (GNNs). Despite this advancement, the issue of social inconsistency is often overlooked, leading to the proposal of a method that samples consistent neighbors and utilizes relation attention mechanisms to tackle this problem effectively. Experiments on real-world datasets validate the effectiveness of the model.

SIGIR '21 - PROCEEDINGS OF THE 44TH INTERNATIONAL ACM SIGIR CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION RETRIEVAL (2021)

Add to Collection

Article Computer Science, Artificial Intelligence