|
An ontology-based system for chronic tropical diseases using the Prot?g?-OWL tool |
|---|---|
| รหัสดีโอไอ | |
| Creator | 1. Segun Olatinwo 2. Justice Emuoyibofarhe 3. Femi Alamu 4. Olusegun Omitola 5. Damilola Olatinwo |
| Title | An ontology-based system for chronic tropical diseases using the Prot?g?-OWL tool |
| Publisher | Research and Development Office, Prince of Songkla University |
| Publication Year | 2561 |
| Journal Title | Songklanakarin Journal of Science and Technology |
| Journal Vol. | 40 |
| Journal No. | 6 |
| Page no. | 1386-1395 |
| Keyword | ontology, formal concept analysis, METHONTOLOGY, precision, recall and F-measure |
| URL Website | http://rdo.psu.ac.th/sjstweb/index.php |
| ISSN | 0125-3395 |
| Abstract | This paper presents the development of an ontology for chronic tropical diseases (CTD). As observed in most studies conducted on CTD domain knowledge, the existing ontologies on CTD do not provide a unified structural model for understanding and sharing this domain knowledge which has made it difficult for medical experts to share a common understanding of this domain. This necessitated the need for an ontology-based system to address the challenges. The research data used in this study were obtained from CTD domain experts. Using formal concept analysis, the acquired knowledge was analyzed to obtain the taxonomy entities of CTD. The obtained entities were used to develop the Ontology of Indigenous Chronic Tropical Diseases (OICTD) through the Prot?g? tool. The developed OICTD was evaluated and compared with the existing Ontology for Tropical Diseases Management (OTDM) using precision, recall and F-measure as performance metrics. The analysis of the acquired knowledge yielded the ontology entities of 936 axioms, 12 classes, 713 logical axioms, 29 object properties, and 182 individuals. The developed OICTD showed a better performance with 97% average precision, 100% average recall, and 98% average F-measure in comparison with the existing OTDM which assumed 78% average precision, 83% average recall, and 80% average F-measure. |
