Database Schema Matching using Corpus-based Semantic Similarity and Word Segmentation

In this paper, we present a new method for database schema matching, the problem of identifying elements of two given schemas that correspond to each other. We use two methods based on a large text corpus: one method for determining the semantic similarity of two target words and the other for automatic word segmentation. We present a name-based element-level database schema matching method that exploits the semantic similarity and the word segmentation method. We also use normalized and modified versions of the Longest Common Subsequence string matching algorithm with weight factors to allow for a balanced combination. Our goal is to develop a schema matching method that uses a single property (element name) for matching and achieves a comparable F-measure score with respect to the methods that use multiple properties (element name, text description, data instance, context description). We validate our method with experimental studies, the results of which suggest that the method is a useful addition to the set of existing schema matchers.