MultiEM: Efficient and Effective Unsupervised Multi-Table Entity Matching

TL;DR

MultiEM introduces a scalable unsupervised approach for multi-table entity matching, significantly improving accuracy and efficiency over existing methods in real-world data management scenarios.

Contribution

This paper presents the first effective and efficient unsupervised multi-table entity matching method, addressing a gap in current research.

Findings

MultiEM outperforms existing methods on six benchmark datasets.

MultiEM is scalable and parallelizable for large datasets.

It achieves higher accuracy and efficiency in multi-table EM tasks.

Abstract

Entity Matching (EM), which aims to identify all entity pairs referring to the same real-world entity from relational tables, is one of the most important tasks in real-world data management systems. Due to the labeling process of EM being extremely labor-intensive, unsupervised EM is more applicable than supervised EM in practical scenarios. Traditional unsupervised EM assumes that all entities come from two tables; however, it is more common to match entities from multiple tables in practical applications, that is, multi-table entity matching (multi-table EM). Unfortunately, effective and efficient unsupervised multi-table EM remains under-explored. To fill this gap, this paper formally studies the problem of unsupervised multi-table entity matching and proposes an effective and efficient solution, termed as MultiEM. MultiEM is a parallelable pipeline of enhanced entity representation, table-wise hierarchical merging, and density-based pruning. Extensive experimental results on six real-world benchmark datasets demonstrate the superiority of MultiEM in terms of effectiveness and efficiency.