Seminar from Mohamed Ettaleb, post-doctoral fellowship at LIUM
Localization: IC2 Classroom, online
Speakert: Mohamed Ettaleb
A recommendation approach based on data mining and multi-layer labeled graphs
The objective of a recommendation system is to assist users in choosing relevant items from a large set of elements. In the current context of the boom in the number of academic publications available (books, articles, etc.) online, providing a personalized recommendation service becomes a necessity.
Moreover, automatic book recommendation based on a query is an emerging theme with many scientific locks. It combines several problems related to information retrieval and data mining for the estimation of the opportunity to recommend a book. This estimation has to be done taking into account the query but also the user’s profile (reading history, interests, notes and comments associated with previous readings) and the whole collection to which the document belongs.
Two main approaches have been addressed in this thesis to deal with the problem of automatic book recommendation: (1) Query-based user intent identification: in this path, we present a framework for automatic user intent identification based on query analysis. (2) Recommending relevant books according to user needs: In the second path, we address complex queries expressed within natural language queries to improve the understanding of user needs within the social recommendation system.