Temporal knowledge graph recommendation with sequence-aware and path reasoning

doi:10.1016/j.datak.2025.102522

Temporal knowledge graph recommendation with sequence-aware and path reasoning

2025-10-03

0

PRE

AI

DOI：10.1016/j.datak.2025.102522

Original

Original request for help

Save

Abstract

En 中文

• This paper presents a TKG that integrates interaction times, interaction relationships, user attributes, and item attributes, comprehensively capturing both the temporal and structural characteristics of user-item interactions, and providing a more complete data model for recommendation. • This paper proposes a novel TKGRec model that integrates sequence-aware and path reasoning. The sequence-aware module employs a dual-attention mechanism to extract temporal variation features of interactions, while the path reasoning module utilizes reinforcement learning to extract relational features of interaction paths, enabling a comprehensive representation of user preferences.

Temporal knowledge graph recommendation with sequence-aware and path reasoning

Abstract

Journal

Organization