Attr-Int: A Simple and Effective Entity Alignment Framework for Heterogeneous Knowledge Graphs

TL;DR

This paper introduces Attr-Int, a straightforward and effective framework for entity alignment in heterogeneous knowledge graphs, addressing structural differences and improving performance over existing methods.

Contribution

The paper proposes a novel attribute interaction framework that enhances entity alignment in heterogeneous KGs and introduces new benchmarks simulating real-world scenarios.

Findings

Attr-Int outperforms state-of-the-art methods on new benchmarks.

The framework effectively integrates attribute information with embedding encoders.

Experiments validate the robustness of Attr-Int across diverse heterogeneous KGs.

Abstract

Entity alignment (EA) refers to the task of linking entities in different knowledge graphs (KGs). Existing EA methods rely heavily on structural isomorphism. However, in real-world KGs, aligned entities usually have non-isomorphic neighborhood structures, which paralyses the application of these structure-dependent methods. In this paper, we investigate and tackle the problem of entity alignment between heterogeneous KGs. First, we propose two new benchmarks to closely simulate real-world EA scenarios of heterogeneity. Then we conduct extensive experiments to evaluate the performance of representative EA methods on the new benchmarks. Finally, we propose a simple and effective entity alignment framework called Attr-Int, in which innovative attribute information interaction methods can be seamlessly integrated with any embedding encoder for entity alignment, improving the performance of existing entity alignment techniques. Experiments demonstrate that our framework outperforms the state-of-the-art approaches on two new benchmarks.