Combining User-Based and Item-Based Collaborative Filtering Using Machine Learning

Collaborative filtering (CF) is typically used for recommending those items to a user which other like-minded users preferred in the past. User-based collaborative filtering (UbCF) and item-based collaborative filtering (IbCF) are two types of CF with a common objective of estimating target user's rating for the target item. This paper explores different ways of combining predictions from UbCF and IbCF with an aim of minimizing overall prediction error. In this paper, we propose an approach for combining predictions from UbCF and IbCF through multiple linear regression (MLR) and support vector regression (SVR). Results of the proposed approach are compared with the results of other fusion approaches. The comparison demonstrates the superiority of the proposed approach. All the tests are performed on a large publically available dataset.