Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
487897 | Procedia Computer Science | 2013 | 8 Pages |
In recent years, several approaches have been proposed to solve the problem of ontology learning. In most approaches, the text representation is only based on the information contained in term weighting and does therefore not process the semantic contained in the sequence in which the words appear. Moreover, the use of many dimensions adds unnecessary noise in the generated model and affects the quality of learning (generalization). Hence, in the present study, we propose a semi-automatic approach that uses the variables selection and clustering to find the candidate changes. In order to identify the correspondence between the ontological artifacts and candidate changes, we used an alignment process. Our approach exploits natural language processing, indexation and machine learning techniques to increase the productivity of ontology engineering task during the enrichment of conceptual model. Good experimental studies demonstrate the multidisciplinary applications of our approach.