Generating core domain ontologies from normalized dictionaries

This paper proposes a general framework for automatic core domain ontology generation from LMF (ISO 24613) standardized dictionaries. The originality of this work lies not only in the use of a unique and finely structured source containing multi-domain and lexical knowledge of morphological, syntactic and semantic levels, lending itself to ontological interpretations, but also in the proper building of the taxonomic backbone of the domain ontology. To this end, we have integrated a validation stage into the proposed process in order to maintain the consistency of the resulting formalized domain ontology core throughout this process and support the checking of anomalies in the handled source. Furthermore, this generation process has been implemented in an iterative and incremental system based on domain- and language-independent rules. The reliability of the proposed process is proven through many experiments that have been conducted on various domains using normalized dictionaries, but without lack of generality, we choose to illustrate an experiment carried out on the Arabic language. This choice is explained by both the great deficiency of work on building of Arabic ontologies and the availability within our research team of an LMF-standardized Arabic dictionary.