Hybrid Customized Query Formulation for Adaptive Information Retrieval

The emergence of new technologies and new users' behaviour that become increasingly demanding. Furthermore, the transformation of user profile: moving from a passive information receiver to an active information generator (social networks, blogs...) as well as other factors have contributed to a radical evolution in the field of information retrieval. Customized Information retrieval has become the new trend that tries to respond in a relevant and satisfying way to the specific users' needs. In this paper, we present a hybrid query formulation: an automatic query generation using (Convolutional Neural Network and Recurrent Neural Network), then an interactive query formulation by the user intervention. We will present the impact of customization at the level of query reformulation on relevance enhancement in the information retrieval process.