Unsupervised Entity Alignment for Temporal Knowledge Graphs

TL;DR

This paper introduces DualMatch, an unsupervised method that effectively fuses relational and temporal information for entity alignment in temporal knowledge graphs, outperforming existing methods without requiring seed alignments.

Contribution

The paper proposes DualMatch, a novel unsupervised approach that encodes and fuses temporal and relational data for entity alignment in TKGs using graph matching.

Findings

Outperforms state-of-the-art in H@1 by 2.4%-10.7%

Achieves higher MRR by 1.7%-7.6%

Works effectively with or without supervision

Abstract

Entity alignment (EA) is a fundamental data integration task that identifies equivalent entities between different knowledge graphs (KGs). Temporal Knowledge graphs (TKGs) extend traditional knowledge graphs by introducing timestamps, which have received increasing attention. State-of-the-art time-aware EA studies have suggested that the temporal information of TKGs facilitates the performance of EA. However, existing studies have not thoroughly exploited the advantages of temporal information in TKGs. Also, they perform EA by pre-aligning entity pairs, which can be labor-intensive and thus inefficient. In this paper, we present DualMatch which effectively fuses the relational and temporal information for EA. DualMatch transfers EA on TKGs into a weighted graph matching problem. More specifically, DualMatch is equipped with an unsupervised method, which achieves EA without necessitating seed alignment. DualMatch has two steps: (i) encoding temporal and relational information into embeddings separately using a novel label-free encoder, Dual-Encoder; and (ii) fusing both information and transforming it into alignment using a novel graph-matching-based decoder, GM-Decoder. DualMatch is able to perform EA on TKGs with or without supervision, due to its capability of effectively capturing temporal information. Extensive experiments on three real-world TKG datasets offer the insight that DualMatch outperforms the state-of-the-art methods in terms of H@1 by 2.4% - 10.7% and MRR by 1.7% - 7.6%, respectively.