Semantic Indexing of Technical Documentation

This research takes place in an industrial context: the CONTINEW Company. This company ensures the storage and security of critical data and technical documentation. Consequently, it is necessary to organize these documents in order to retrieve quickly critical information. The management of this increasing volume of documents requires document classification which is based on indexing techniques. So, how much relevant the indexing phase is, more relevant the classification will be. The technical documentation is by nature strongly structured. For example, the logical structure describes the role and the nature of the document elements (introduction, title, section, and so one...) and the logical links between them (A chapter is composed of section and so one). Such structure facilitates document presentation and improves the indexing precision. The classical information retrieval systems use neither the logical structure, nor the concept contained in the textual content of documents. The document semantic is described by concepts belonging to a semantic resource. In this context, we propose a new semantic indexing model which exploits both the logical structures and the semantic contents of documents.