Ordinary User Oriented Model Construction for Assisting Conversational Agents

In this paper, we defend the idea of integrating the actual cognitive features of novice users in the creation process of the applications to significantly increase the quality of the function of assistance. First we present a framework dedicated to the recording of novice user natural language utterances in assisting situations. Then we propose a first analysis of the collected corpus in terms of the perceptual features appearing in the linguistic referential expressions