A recommendation system (or recommender) is an algorithm whose goal is to recommend products to potential users. To achieve its task, it uses information about some user preferences. We present recommenders that use information about the preferences of only a very small subset of users (called a committee) on a very small set of products called the witness products set. The main interest of our approach compared to previous ones is that it needs substantially less data for ensuring a very good quality of recommendation. © 2010 Springer-Verlag Berlin Heidelberg.
CITATION STYLE
Hémon, S., Largillier, T., & Peyronnet, S. (2010). Partial ranking of products for recommendation systems. In Lecture Notes in Business Information Processing (Vol. 61 LNBIP, pp. 265–277). Springer Verlag. https://doi.org/10.1007/978-3-642-15208-5_24
Mendeley helps you to discover research relevant for your work.