This research takes place in an industrial context: the CONTINEW Company. This company ensures the storage and security of critical data and technical documentation. The term technical documentation refers to different documents with product-related data and information that are used and stored for different purposes, such as user manuals and product specifications. They are strongly structured, but different authors have used different styles and models for document construction. The management of this increasing volume of documents requires document classification in order to retrieve information quickly and to construct a standard model for each category of documents.