Collaborative filtering based on content addressing

Collaborative Filtering (CF) is one of the most popular recommendation techniques. It is based on the assumption that users with similar tastes prefer similar items. One of the major drawbacks of the CF is its limited scalability, as the complexity of the CF grows linearly both with the number of available users and items. This work proposes a new fast variant of the CF employed over multi-dimensional content-addressable space. Our approach heuristically decreases the computational effort required by the CF algorithm by limiting the search process only to potentially similar users. Experimental results demonstrate that our approach is capable of generate recommendations with high levels of accuracy, while significantly improving performance in comparison with the traditional implementation of the CF.