Ontology-Based User Preferences and Social Search for Spoken Dialogue Systems

Many current spoken dialogue systems for search are domain-specific and do not take into account the preferences and interests of the user. In order to provide a more personalized answer tailored to the user needs, we propose a spoken dialogue system where user interests are expressed as scores in modular ontologies. This also allows us to cover multiple domains (e.g. searching for restaurant, housing,...) because each ontology module corresponds to a search domain. This approach allows for a dynamic and evolving representation of user interests. Moreover, a collaborative search of users with similar interests allows to build ad-hoc communities where information can be shared amongst and recommended to users. We propose to use techniques borrowed from formal concept analysis to flexibly and efficiently build these communities.