Abstract:
The Semantic Web is the salient technology of knowledge management, consisting of data extraction and annotation processes, which requires semantic representation to express data in an ontological format. The ontological extraction of unstructured data to enable the automatic generation of concepts and relations has led us to the presentation of our unique approach of automatic ontology extraction. However, domain experts are still required to modify the structure of ontological results, which makes the process very timeconsuming and costly. Yet, there still exists the need for an ontology-based semantic extraction approach from text corpus to discover concepts, instances, and semantic relations between concepts or instances. This paper presents an approach of an ontology-based semantic extraction and the accompanying semantic extraction rules, as applied to tourism domain. The proposed semantic extraction rules are defined as extension rules working with GATE API. As a result, the efficiency of the proposed ontological extraction approach is validated through the Precision, Recall and F-measure scores, with average values of 91.48%, 89.12%, and 90.23%, respectively.
Page(s):
2721-2731
DOI:
DOI not available
Published:
Journal: Journal of Theoretical and Applied Information Technology, Volume: 98, Issue: 14, Year: 2020