An Approach Based on Patterns for Synonymy Relations Detection. Issue 1 (February 2021)

• Record Type: Journal Article
• Authors: Tovar, M; Flores-Petlacalco, G; Reyes-Ortiz, J A
• Is Part Of: Journal of physics. Volume 1828:Issue 1(2021)
• Publication Date: 2021-02
• Subjects: Physics -- Congresses; 530.5
• DOI: 10.1088/1742-6596/1828/1/012008
• Languages: English
• ISSNs: 1742-6588

Abstract

Abstract: Thesaurus, taxonomies, and ontologies are created as a way to represent knowledge. In recent years, the efforts to create them are focused on automatic knowledge extraction. For this purpose, research in the extraction of concepts and semantic relations is needed. Concepts or entities extracted from unstructured data has a became a challenge for the research communities. Texts, specifically, scientific papers contain knowledge where experts express their contributions to the science, including: techniques, methods and claims. In order to describe them, experts use synonyms, which are relevant to consider whereas it is desirable to extract such type of contribution form scientific text. This paper describes a method based on patterns for semantic relations extraction from scientific texts. The semantic relations correspond to synonymy. We describe pattern acquisition and relations extraction as phases to data evaluation. The evaluation shows promising results with 75% precision, 58% recall and 65% F-measure.