An Industry Evaluation of Embedding-based Entity Alignment

TL;DR

This paper evaluates embedding-based entity alignment methods in industrial settings, analyzing their performance with various seed mapping sizes and biases, and introduces a new medical knowledge graph benchmark.

Contribution

It provides an industrial evaluation of state-of-the-art methods and introduces a new benchmark from medical knowledge graphs for practical assessment.

Findings

Performance varies with seed mapping size and bias

Existing methods have specific advantages and disadvantages in industrial contexts

A new medical knowledge graph benchmark was created and evaluated.

Abstract

Embedding-based entity alignment has been widely investigated in recent years, but most proposed methods still rely on an ideal supervised learning setting with a large number of unbiased seed mappings for training and validation, which significantly limits their usage. In this study, we evaluate those state-of-the-art methods in an industrial context, where the impact of seed mappings with different sizes and different biases is explored. Besides the popular benchmarks from DBpedia and Wikidata, we contribute and evaluate a new industrial benchmark that is extracted from two heterogeneous knowledge graphs (KGs) under deployment for medical applications. The experimental results enable the analysis of the advantages and disadvantages of these alignment methods and the further discussion of suitable strategies for their industrial deployment.