An object-oriented approach for an adaptive interface with dual-classification of the user

The focus of our work is to provide users with an optimal working environment for retrieval from databases, without its intervention for any modification. To do this, we develop an adaptive interface supported by user models represented as stereotypes. In an object-oriented approach, we create two hierarchies of stereotypes (taxonomies of classes). They class the user according to criteria respectively independent and dependent of the database domain, allowing to adapt the syntax and the semantics of the user-database interaction. The user works in a fitted environment and obtains adapted answers to his global need