An Indexing Approach based on a Hybrid Model of Terminology-extraction using a Filtering by Elimination Terms

The extraction of terms is an important step in building a resource for indexing and many powerful tools are available for several languages. This complex process, which identifies candidate terms may become indexes for annotations or documents, is often subject to the problem of lack of relevance of calculated terms. Consequently, the extraction of terminology from the texts must be strong and solid to handle the errors and suggest better results, without encumbering the user with too many proposals of indexes. It is, in this perspective that we propose here a new indexing approach based on a hybrid model of terminologies extraction using a filtering by elimination terms and which operates on a corpus of annotated images with legends . Index Terms-term extraction, semantic indexing, linguistic analysis, syntactic patterns, complex terms, n-grams .