Filtragem Colaborativa Incremental para recomendações automáticas na Web

摘要: The use of collaborative filtering recommenders on the Web is typically done in environments where data constantly flowing and new customers products are emerging. In this work, it proposed an incremental version item-based Collaborative Filtering for implicit binary ratings. It compared with a non-incremental one, as well user-based approach. also study techniques working sparse matrices these algorithms. All versions implemented R empirically evaluated five different datasets various number users and/or items. observed that measure Recall used tend to improve when we continuously add information recommender model time spent recommendation does not degrade. Time updating similarity matrix (necessary recommendation) relatively low motivates