Syntactic Enrichment of LMF Normalized Dictionaries Based on the Context-Field Corpus

In this paper, we deal with the representation of syntactic knowledge, particularly the syntactic behavior of verbs. In this context, we propose an approach to identify syntactic behaviors from a corpus based on the LMF Context-Field in order to enrich the syntactic extension of LMF normalized dictionary. Our approach consists of the following steps: (i) Identification of syntactic patterns, (ii) Construction of a grammar suitable for each syntactic pattern, (iii) Construction of a corpus from the LMF normalized dictionary, (iv) Application of grammars to the corpus and (v) Enrichment of the LMF dictionary. To validate this approach, we carried out an experiment that focuses on the syntactic behavior of Arabic verbs. We used the NooJ linguistic platform and an available LMF Arabic dictionary that contains 37,000 entries and 10,800 verbs. The obtained results concerning more than 7,800 treated verbs show 85 % of precision and 87 % of recall.