Extracting Attentive Social Temporal Excitation for Sequential Recommendation

TL;DR

This paper introduces STEN, a novel time-aware sequential recommendation framework that models fine-grained social temporal influences on user behavior using temporal point processes, improving recommendation accuracy and explainability.

Contribution

It proposes a direct event-level social temporal modeling approach with temporal point processes, decomposing effects into social and ego influences, and demonstrates superior performance over existing methods.

Findings

STEN outperforms baseline methods on three real-world datasets.

The model provides interpretable, event-level recommendation explanations.

Temporal mutual and ego effects are effectively captured and utilized.

Abstract

In collaborative filtering, it is an important way to make full use of social information to improve the recommendation quality, which has been proved to be effective because user behavior will be affected by her friends. However, existing works leverage the social relationship to aggregate user features from friends' historical behavior sequences in a user-level indirect paradigm. A significant defect of the indirect paradigm is that it ignores the temporal relationships between behavior events across users. In this paper, we propose a novel time-aware sequential recommendation framework called Social Temporal Excitation Networks (STEN), which introduces temporal point processes to model the fine-grained impact of friends' behaviors on the user s dynamic interests in an event-level direct paradigm. Moreover, we propose to decompose the temporal effect in sequential recommendation into social mutual temporal effect and ego temporal effect. Specifically, we employ a social heterogeneous graph embedding layer to refine user representation via structural information. To enhance temporal information propagation, STEN directly extracts the fine-grained temporal mutual influence of friends' behaviors through the mutually exciting temporal network. Besides, the user s dynamic interests are captured through the self-exciting temporal network. Extensive experiments on three real-world datasets show that STEN outperforms state-of-the-art baseline methods. Moreover, STEN provides event-level recommendation explainability, which is also illustrated experimentally.