Sematch: Semantic similarity framework for Knowledge Graphs

Abstract

Sematch is an integrated framework for the development, evaluation and application of semantic similarity for Knowledge Graphs. The framework provides a number of similarity tools and datasets, and allows users to compute semantic similarity scores of concepts, words, and entities, as well as to interact with Knowledge Graphs through SPARQL queries. Sematch focuses on knowledge-based semantic similarity that relies on structural knowledge in a given taxonomy (e.g. depth, path length, least common subsumer), and statistical information contents. Researchers can use Sematch to develop and evaluate semantic similarity metrics and exploit these metrics in applications.