An Improved Collaborative Filtering Recommendation Algorithm Incorporating Opinions Analysis

Collaborative filtering recommendation algorithm has become a common way to deal with the problem of information overload, which hinders consumers to make appropriate decisions and firms to provide the items that consumers really interest in. Traditional collaborative method is basing on consumers' rating on the items, hence, their performance suffers from data sparsity and cold-start. In this paper, we propose the framework of a novel recommendation algorithm. The proposed algorithm adopts the method of opinion mining to extract consumers' preference from their reviews, and then incorporating it to collaborative filtering method to improve the performance of the algorithm. The current work is an improving method to the traditional item-based collaborative filtering algorithm.