Towards detecting anomalies in the content of standardized LMF dictionaries

Dictionaries are reference resources for learning and diffusing natural languages. Their contents must be enriched carefully due to their importance. However, such contents might contain errors and inconsistencies that are hard to detect manually. Several researches have been made in recent years in order to perform this step automatically. However, they have dealt with the problem in a superficial way. The present paper deals with the detection of anomalies in the content of LMF-standardized dictionaries that covers lexical knowledge at the morphological, syntactic and semantic levels. Thus, we are proposing an approach based on a typological study of the potential anomalies that can occur in editorial dictionaries in general. This approach takes advantage of the LMF fine structure that highlights all kinds of relationships between entries’ knowledge and distinguishes the role of each available text such as giving definitions and examples. An experiment of the proposed approach was carried out on an available LMFstandardized dictionary of the Arabic language. This experiment has been related to the morphological and syntactic levels.