El idrissi esserhrouchni, O. and Frikh, B. and Ouhbi, B. (2016) Ontologyline: A new framework for learning non-taxonomic relations of domain ontology. Communications in Computer and Information Science, 631. pp. 345-364.

Full text not available from this repository.
Official URL: https://www.scopus.com/inward/record.uri?eid=2-s2....

Abstract

Domain Ontology learning has been introduced as a technology that aims at reducing the bottleneck of knowledge acquisition in the construction of domain ontologies. However, the discovery and the labelling of non-taxonomic relations have been identified as one of the most difficult problems in this learning process. In this paper, we propose OntologyLine, a new system for discovering non-taxonomic relations and building domain ontology from scratch. The proposed system is based on adapting Open Information Extraction algorithms to extract and label relations between domain concepts. OntologyLine was tested in two different domains: the financial and cancer domains. It was evaluated against gold standard ontology and was compared to state-of-the-art ontology learning algorithm. The experimental results show that OntologyLine is more effective for acquiring non-taxonomic relations and gives better results in terms of precision, recall and F-measure. © Springer International Publishing AG 2016.

Item Type: Article
Uncontrolled Keywords: Diseases; Information analysis; Information retrieval; Knowledge acquisition; Knowledge engineering; Knowledge management; Learning algorithms; Mergers and acquisitions; Ontology, Different domains; Domain concepts; Domain ontologies; F measure; Gold standards; Learning process; Ontology learning; State of the art, Data mining
Subjects: Computer Science
Divisions: SCIENTIFIC PRODUCTION > Computer Science
Depositing User: Administrateur Eprints Administrateur Eprints
Last Modified: 31 Jan 2020 15:46
URI: http://eprints.umi.ac.ma/id/eprint/2631

Actions (login required)

View Item View Item