Research Article
Dual-Channel Graph Contextual Self-Attention Network for Session-Based Recommendation
@INPROCEEDINGS{10.1007/978-3-030-92635-9_15, author={Teng Huang and Huiqun Yu and Guisheng Fan}, title={Dual-Channel Graph Contextual Self-Attention Network for Session-Based Recommendation}, proceedings={Collaborative Computing: Networking, Applications and Worksharing. 17th EAI International Conference, CollaborateCom 2021, Virtual Event, October 16-18, 2021, Proceedings, Part I}, proceedings_a={COLLABORATECOM}, year={2022}, month={1}, keywords={Session-based recommendation Graph neural network Self-attention network}, doi={10.1007/978-3-030-92635-9_15} }- Teng Huang
Huiqun Yu
Guisheng Fan
Year: 2022
Dual-Channel Graph Contextual Self-Attention Network for Session-Based Recommendation
COLLABORATECOM
Springer
DOI: 10.1007/978-3-030-92635-9_15
Abstract
The session-based recommendation task is a key task in many online service websites (such as online music, e-commerce, etc.). Its goal is to predict the user’s next possible interactive item based on an anonymous user behavior sequence. However, existing methods do not take into account the information implicit in similar neighbor sessions. This paper proposes a new dual-channel graph neural network combined with a self-attention network model. In this model, the graph neural network is used to model the session sequence, and the learning is different through two independent channels between and within the session. The model uses the attention network to learn the weights of different items for the recommendation results. Several experiments are done on two public e-commerce data sets and the results show that the performance of the model proposed in this paper is better than the general methods.
