Efficient Taxonomic Similarity Joins with Adaptive Overlap Constraint

TL;DR

This paper introduces a novel similarity join method that leverages taxonomy knowledge to identify semantically similar records efficiently, using an adaptive overlap constraint and optimized prefix filtering to enhance performance.

Contribution

It presents a new similarity measure based on taxonomy relations and an adaptive prefix filtering algorithm that maximizes filtering power through parameter optimization.

Findings

Outperforms state-of-the-art methods significantly in efficiency

Demonstrates high scalability on large datasets

Achieves better semantic matching accuracy

Abstract

A similarity join aims to find all similar pairs between two collections of records. Established approaches usually deal with synthetic differences like typos and abbreviations, but neglect the semantic relations between words. Such relations, however, are helpful for obtaining high-quality joining results. In this paper, we leverage the taxonomy knowledge (i.e., a set of IS-A hierarchical relations) to define a similarity measure which finds semantic-similar records from two datasets. Based on this measure, we develop a similarity join algorithm with prefix filtering framework to prune away irrelevant pairs effectively. Our technical contribution here is an algorithm that judiciously selects critical parameters in a prefix filter to maximise its filtering power, supported by an estimation technique and Monte Carlo simulation process. Empirical experiments show that our proposed methods exhibit high efficiency and scalability, outperforming the state-of-art by a large margin.