Related Communities:

Applying of Machine Learning Techniques to Combine String-based, Language-based and Structure-based Similarity Measures for Ontology Matching

Applying of Machine Learning Techniques to Combine String-based, Language-based and Structure-based Similarity Measures for Ontology Matching

Author(s): Lev Bulygin, Sergey Stupnikov
Published:Data Analytics and Management in Data Intensive Domains: I International Conference DADID/RCDL' 2019 (October 1518, 2019, Kazan, Russia): Conference Proceedings. Edited b Alexander Elizarov, Boris Novikov, Sergey Stupnikov. Selected Papers of the XXI International Conference on Data Analytics and Management in Data Intensive Domains (DAMDID/RCDL 2019). CEUR Workshop Proceedings, Vol. 2523. P. 129-147. 2019
Abstract:
In the areas of Semantic Web and data integration, ontology matching is one of the important steps to resolve semantic heterogeneity. Manual ontology matching is very labor-intensive, time-consuming and prone to errors. So development of automatic or semi-automatic ontology matching methods and tools is quite important. This paper applies machine learning with different similarity measures between ontology elements as features for ontology matching. An approach to combine string-based, language-based and structurebased similarity measures with machine learning techniquesis proposed. Logistic Regression, Random Forest classifier and Gradient Boosting are used as machine learning methods. The approach is evaluated on two datasets of Ontology Alignment Evaluation Initiative (OAEI).
Download: [ http://ceur-ws.org/Vol-2523/paper14.pdf ]

Supported by Synthesis Group