Historically Relevant Event Structuring for Temporal Knowledge Graph Reasoning

2024-05-17

arXiv.org

AI Summary

Existing temporal knowledge graph (TKG) reasoning models struggle to capture multi-granular temporal interactions among recent snapshots and overlook semantically rich, long-horizon-relevant historical events strongly associated with the query—leading to insufficient modeling of historical dependencies and evolutionary trends. To address this, we propose a structured historical relevance modeling framework: a multi-granular evolution encoder captures local temporal dynamics, while a global relevance encoder identifies critical cross-temporal links; an adaptive gating mechanism fuses these complementary signals. Our approach integrates multi-granular graph neural networks, global attention, temporal positional encoding, and structured representation learning. Evaluated on four event-centric benchmark datasets, our method achieves state-of-the-art performance, significantly improving future event prediction accuracy. Results demonstrate both the effectiveness and generalizability of structured historical relevance modeling.

Technology Category

• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Reasoning under Uncertainty: Graphical Models
• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Web Mining and Content Analysis: Models for Web evolution

Abstract

Temporal Knowledge Graph (TKG) reasoning focuses on predicting events through historical information within snapshots distributed on a timeline. Existing studies mainly concentrate on two perspectives of leveraging the history of TKGs, including capturing evolution of each recent snapshot or correlations among global historical facts. Despite the achieved significant accomplishments, these models still fall short of I) investigating the impact of multi-granular interactions across recent snapshots, and II) harnessing the expressive semantics of significant links accorded with queries throughout the entire history, particularly events exerting a profound impact on the future. These inadequacies restrict representation ability to reflect historical dependencies and future trends thoroughly. To overcome these drawbacks, we propose an innovative TKG reasoning approach towards extbf{His}torically extbf{R}elevant extbf{E}vents extbf{S}tructuring (HisRES). Concretely, HisRES comprises two distinctive modules excelling in structuring historically relevant events within TKGs, including a multi-granularity evolutionary encoder that captures structural and temporal dependencies of the most recent snapshots, and a global relevance encoder that concentrates on crucial correlations among events relevant to queries from the entire history. Furthermore, HisRES incorporates a self-gating mechanism for adaptively merging multi-granularity recent and historically relevant structuring representations. Extensive experiments on four event-based benchmarks demonstrate the state-of-the-art performance of HisRES and indicate the superiority and effectiveness of structuring historical relevance for TKG reasoning.

Problem

• Investigates multi-granular interactions in recent TKG snapshots
• Harnesses expressive semantics of significant historical event links
• Overcomes limitations in reflecting historical dependencies and future trends

Innovation

• Multi-granularity evolutionary encoder captures snapshot dependencies
• Global relevance encoder focuses on crucial event correlations
• Self-gating mechanism adaptively merges historical representations