Authors:
B. Frikh
1
;
A. S. Djaanfar
2
and
B. Ouhbi
3
Affiliations:
1
Ecole Supérieure de Technologie, Morocco
;
2
Dhar El Mahraz, Morocco
;
3
Ecole Nationale Superieure d’Arts et Métiers, Morocco
Keyword(s):
Domain ontology, CHIR-statistic, Mutual information, Hybrid approach, Web mining.
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence
;
Context
;
Data Engineering
;
Enterprise Information Systems
;
Information Systems Analysis and Specification
;
Knowledge Acquisition
;
Knowledge Engineering and Ontology Development
;
Knowledge Reengineering
;
Knowledge Representation
;
Knowledge-Based Systems
;
Ontologies and the Semantic Web
;
Ontology Engineering
;
Paradigm Trends
;
Software Engineering
;
Symbolic Systems
Abstract:
This paper describes a hybrid statistical and semantic relationships among model concepts for ontology construction. The implementation of the model, called HCHIRSIM (Hybrid Chir-Statistic and Similarity), can be adapted to any domain ontology learning from the Web. It can be viewed as a combination of information from inference view of concepts by using the CHIR-statistic method and the semantic relationships among concepts from the Web by the mutual information measure. The experiments show that our hybrid approach outperforms both purely statistical and purely semantic relationships among concepts approaches. The successful evaluation of our method with different values of the weighting parameter shows that the proposed approach can effectively construct a cancer domain ontology from unstructured text documents.