Instance Based Schema Matching Framework Utilizing Google Similarity and Regular Expression

Abstract

Schema matching is the task of identifying correspondences between schema attributes that exist in different schemas. A variety of approaches have been proposed to achieve the main goal of high-quality match results with respect to precision (P) and recall (R). However, these approaches are unable to achieve high quality match results, as most of these approaches treated the instances as string regardless the data types of the instances. As a consequence, this causes unidentii??ed matches especially for attribute with numeric instances which further reduces the quality of match results. Therefore, effort still needs to be done to further improve the quality of the match results. In this paper, we propose a framework for addressing the problem of finding matches between schemas of semantically and syntactically related data. Since we only fully exploit the instances of the schemas for this task, we rely on strategies that combine the strength of Google as a web semantic and regular expression as pattern recognition. To demonstrate the accuracy of our framework, we conducted an experimental evaluation using real world data sets. The results show that our framework is able to find 1-1 schema matches with high accuracy in the range of 93% - 99% in terms of precision (P), recall (R), and F-measure (F).