Session-based Recommendation Based on Long-term and Short-term Interest Incorporating Social Information

In session-based recommendation systems, user interests are dynamic and purchasing actions are often influenced by both long-term and short-term interest preferences. However, user interests are not solely influenced by the users themselves but also by other external factors, such as social connections. To address these issues, a method utilizing a Graph Attention Network is proposed, which effectively integrates both the long-term and short-term interests of users and their friends. Experiments on the Douban and Delicious datasets demonstrate that the proposed algorithm outperforms baseline models.