Aligning Cross-Lingual Entities with Multi-Aspect Information

TL;DR

This paper presents a novel embedding-based approach combining graph convolutional networks and multilingual BERT to improve cross-lingual entity alignment in knowledge graphs, achieving state-of-the-art results.

Contribution

It introduces a multi-aspect GCN and BERT integration strategy for better cross-lingual entity embedding and alignment, outperforming previous methods.

Findings

Significant performance improvements over existing systems.

Effective use of multi-aspect information and multilingual BERT.

Robust results on benchmark datasets.

Abstract

Multilingual knowledge graphs (KGs), such as YAGO and DBpedia, represent entities in different languages. The task of cross-lingual entity alignment is to match entities in a source language with their counterparts in target languages. In this work, we investigate embedding-based approaches to encode entities from multilingual KGs into the same vector space, where equivalent entities are close to each other. Specifically, we apply graph convolutional networks (GCNs) to combine multi-aspect information of entities, including topological connections, relations, and attributes of entities, to learn entity embeddings. To exploit the literal descriptions of entities expressed in different languages, we propose two uses of a pretrained multilingual BERT model to bridge cross-lingual gaps. We further propose two strategies to integrate GCN-based and BERT-based modules to boost performance. Extensive experiments on two benchmark datasets demonstrate that our method significantly outperforms existing systems.