Collective Entity Alignment via Adaptive Features

TL;DR

This paper introduces a collective entity alignment framework that leverages structural, semantic, and string features, formulated as a stable matching problem, to improve accuracy in aligning entities across knowledge graphs.

Contribution

It proposes a novel collective EA approach using adaptive features and stable matching, addressing interdependence among entities ignored by prior methods.

Findings

Outperforms state-of-the-art solutions on cross-lingual benchmarks

Effective in both cross-lingual and mono-lingual settings

Demonstrates superiority through empirical evaluation

Abstract

Entity alignment (EA) identifies entities that refer to the same real-world object but locate in different knowledge graphs (KGs), and has been harnessed for KG construction and integration. When generating EA results, current solutions treat entities independently and fail to take into account the interdependence between entities. To fill this gap, we propose a collective EA framework. We first employ three representative features, i.e., structural, semantic and string signals, which are adapted to capture different aspects of the similarity between entities in heterogeneous KGs. In order to make collective EA decisions, we formulate EA as the classical stable matching problem, which is further effectively solved by deferred acceptance algorithm. Our proposal is evaluated on both cross-lingual and mono-lingual EA benchmarks against state-of-the-art solutions, and the empirical results verify its effectiveness and superiority.