Multi-Aggregator Time-Warping Heterogeneous Graph Neural Network for Personalized Micro-Video Recommendation

2022-10-17

International Conference on Information and Knowledge Management

AI Summary

To address the insufficient modeling of timeliness and the difficulty in capturing dynamic user interest evolution in news-oriented micro-video recommendation, this paper proposes a session-level graph neural network method that jointly incorporates temporal awareness and heterogeneous interaction. We construct a session-driven heterogeneous graph and design multi-granularity neighbor aggregators alongside a learnable time-warping function to jointly model content recency, interest drift, and interaction heterogeneity. This work is the first to integrate a multi-aggregator coordination mechanism with explicit time warping into a micro-video recommendation framework, enabling unified representation of dynamic preference evolution. Extensive experiments on multiple real-world micro-video datasets demonstrate that our approach achieves a 12.6% improvement in Recall@10 over state-of-the-art methods, validating its effectiveness and superiority for time-sensitive recommendation tasks.

Technology Category

• Machine Learning: Time-Series/Data Streams
• Computer Vision: Video Understanding & Activity Analysis
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Web Mining and Content Analysis: Content-based information diffusion

Abstract

Micro-video recommendation is attracting global attention and becoming a popular daily service for people of all ages. Recently, Graph Neural Networks-based micro-video recommendation has displayed performance improvement for many kinds of recommendation tasks. However, the existing works fail to fully consider the characteristics of micro-videos, such as the high timeliness of news nature micro-video recommendation and sequential interactions of frequently changed interests. In this paper, a novel Multi-aggregator Time-warping Heterogeneous Graph Neural Network (MTHGNN) is proposed for personalized news nature micro-video recommendation based on sequential sessions, where characteristics of micro-videos are comprehensively studied, users' preference is mined via multi-aggregator, the temporal and dynamic changes of users' preference are captured, and timeliness is considered. Through the comparison with the state-of-the-arts, the experimental results validate the superiority of our MTHGNN model.

Problem

• Micro-video Recommendation
• Temporal Dynamics
• User Interest Evolution

Innovation

• MTHGNN
• Personalized Recommendation
• Heterogeneous Graph Neural Network