Collaborative filtering recommendation combining FCM and Slope One algorithm

In view of the data sparseness problem existed in the traditional collaborative filtering recommendation algorithm, this paper proposes a hybrid collaborative filtering recommender framework integrated FCM clustering and Slope One algorithm and FSUBCF algorithm. Firstly this algorithm use the Slope One algorithm based on FCM cluster to predict item ratings that users have not rated in matrix, and then, to implement recommendation by the collaborative filtering recommendation algorithm based on user. The experimental results show that this algorithm can improved the prediction accuracy compared to the original Slope One algorithm and can adapt to the data sparser recommendation system. Compared with other traditional collaborative filtering algorithms, the recommendation accuracy also has obvious advantages.