It's AI Match: A Two-Step Approach for Schema Matching Using Embeddings

TL;DR

This paper introduces a two-step neural embedding-based method for schema matching, significantly improving the accuracy and robustness of identifying semantic correspondences between data schemas, reducing manual effort in data integration.

Contribution

It presents a novel end-to-end schema matching approach using embeddings at both table and attribute levels, outperforming traditional methods in finding complex correspondences.

Findings

Robust and reliable schema matching results.

Ability to identify non-trivial correspondences.

Outperforms traditional schema matching approaches.

Abstract

Since data is often stored in different sources, it needs to be integrated to gather a global view that is required in order to create value and derive knowledge from it. A critical step in data integration is schema matching which aims to find semantic correspondences between elements of two schemata. In order to reduce the manual effort involved in schema matching, many solutions for the automatic determination of schema correspondences have already been developed. In this paper, we propose a novel end-to-end approach for schema matching based on neural embeddings. The main idea is to use a two-step approach consisting of a table matching step followed by an attribute matching step. In both steps we use embeddings on different levels either representing the whole table or single attributes. Our results show that our approach is able to determine correspondences in a robust and reliable way and compared to traditional schema matching approaches can find non-trivial correspondences.