Target Tracking in the recommender space - Toward a new recommender system based on Kalman filtering

Samuel Nowakowski (1) , Anne Boyer (1) , Cédric Bernier (1)

1 KIWI - Knowledge Information and Web Intelligence

We assume that users and their consumptions of television programs are vectors in the multidimensional space of the categories of the resources. Knowing this space, we propose an algorithm based on a Kalman filter to track the user's profile and to foresee the best prediction of their future position in the recommendation space. From this prediction, we build a recommendation of contents.

Domaines: Web

Samuel Nowakowski, Anne Boyer, Cédric Bernier. Target Tracking in the recommender space - Toward a new recommender system based on Kalman filtering. INTERNET 2011, Jun 2011, Luxembourg, Luxembourg. pp.non connus. ⟨hal-00591812⟩