TransClean: Finding False Positives in Multi-Source Entity Matching under Real-World Conditions via Transitive Consistency

We present TransClean, a method for detecting false positive predictions of entity matching algorithms under real-world conditions characterized by large-scale, noisy, and unlabeled multi-source datasets that undergo distributional shifts. TransClean is explicitly designed to operate with multiple data sources in an efficient, robust and fast manner while accounting for edge cases and requiring limited manual labeling. TransClean leverages the Transitive Consistency, a measure that we propose, aimed to detect the lack of consistency of a pairwise matching model f θ on the graph G f θ implied by its predicted pairwise matches. The Transitive Consistency is calculated via all the predictions of f θ on the pairs of records belonging to the same connected components in G f θ. TransClean iteratively modifies a matching through gradually removing false positive matches while removing as few true positive matches as possible. In each of these steps, the estimation of the Transitive Consistency is exclusively done through model inference and produces indicators that can be used as proxies of the amounts of true and false positives in the matching while not requiring any manual labeling. These indicators produce estimates of the quality of the matching and point out which record groups are likely to contain false positives. In our experiments, we compare combining TransClean with a naively trained pairwise matching model (DistilBERT) and with a state-of-the-art end-toend matching method (CLER) and illustrate the flexibility of TransClean in being able to detect most of the false positives of either setup across a variety of datasets. Our experiments show that TransClean induces an average +24.42 F1 score improvement for entity matching in a multi-source setting when compared to traditional pair-wise matching algorithms.