Title: MINING SEMANTICS FROM MEDICAL STRUCTURED CORPUSES TO BUILD AND ENRICH DOMAIN ONTOLOGIES

Issue Number: Vol. 5, No. 2
Year of Publication: 2015
Page Numbers: 95-110
Authors: Driouche Razika
Journal Name: International Journal of Digital Information and Wireless Communications (IJDIWC)
- Hong Kong
DOI:  http://dx.doi.org/10.17781/P001646

Abstract:


Ontologies have become an important means for structuring knowledge, building knowledge-intensive systems and applying to open and dynamic environments wherever the knowledge they model, as well as the user needs which they have to answer evolve. Therefore, the ontologies must be regularly updated at the risk of becoming obsolete. The construction and maintenance of domain ontologies are costly and time-consuming tasks.With the advent of ontologies being used in many different fields of computer science, developing appropriate methods to automate ontology engineering become an increasingly important goal. Hence, we present an extraction, building and enrichment process to generate and enrich domain ontology using text-mining techniques. We exploit domain specific texts collected from medical reports in order to automatically extract terms. These terms are sets of concepts/relations which have either taxonomic or non-taxonomic relationships among them. For the building need, we have developed an extensive and detailed ontology in the field of gynaecology using this process. For the enrichment need, we arrange the extracted terms in sub-hierarchies, and detect the appropriate place to attach them to the ontology hierarchy