Incorporating context into collaborative filtering systems: a survey
Andrei Ponomarev · Informatics and Automation · 2014
It is recognized that incorporating context information into recommender systems is one of the most effective ways to increase their quality and predictive abilities. The paper surveys primary methods of enhancing collaborative filtering systems by taking actual context information into account. The focus is mostly on different flavours of contextual prefiltering and matrix factorization approaches which are the most popular and promising.