Analyses and Fundamental ideas for a Relation Extraction Approach

Relation extraction is a difficult open research problem with important applications in several fields such as knowledge management, web mining, ontology building, intelligent systems, etc. In our research, we focus on extracting relations among the ontological concepts in order to build a domain ontology. In this paper, firstly, we answer some crucial questions related to the text analyses, the word features and the various relation types. Secondly, we use this theoretical analysis and some issues to define the fundamental ideas of our new approach. Our objective is to extract multi-type relations from the text analyses and the existent relations (in the concept hierarchy). Our approach combines a verb centered method, lexical analyses, syntactic and statistic ones. It is based on an exclusive interest to the document style during the statistic process, a rich contextual modelling that strengthens the term cooccurrence selection, a lexical analysis, a use of the existent relations in the concept hierarchy and a stepping between the various extracted relations to facilitate the evaluation made by the domain experts. Thirdly, we present an illustrative example to explain the previous ideas.