Ontology Maintenance using Textual Analysis

Abstract

Ontologies are continuously confronted to evolution problem. Due to the complexity of the changes to be made, a maintenance process, at least a semi-automatic one, is more and more necessary to facilitate this task and to ensure its reliability. In this paper, we propose a maintenance ontology model for a domain, whose originality is to be language independent and based on a sequence of text processing in order to extract highly related terms from corpus. Initially, we deploy the document classification technique using GRAMEXCO to generate classes of texts segments having a similar information type and identify their shared lexicon, agreed as highly related to a unique topic. This technique allows a first general and robust exploration of the corpus. Further, we apply the Latent Semantic Indexing method to extract from this shared lexicon, the most associated terms that has to be seriously considered by an expert to eventually confirm their relevance and thus updating the current ontology. Finally, we show how the complementarity between these two techniques, based on cognitive foundation, constitutes a powerful refinement process.

Keywords

• singular value decomposition
• latent semantic indexing
• correlated terms
• vectorial model
• cognitive psychology
• classification
• ontology maintenance