Abstract:

With the extreme sparsity of the data, traditional collaborative filtering similarity metrics are unable to obtain accurate recommendation results. In order to solve this problem, a collaborative filtering algorithm based on user feature and cloud model is proposed. Firstly, it takes advantage of the cloud model to calculate the similarity of user rating cloud, combines with user scoring preference to fill the original matrix, and then get user cloud similarity. Secondly, combining with user feature similarity and user cloud similarity, a new similarity measure method is proposed to calculate the final similarity by using weighting factor. Finally, the final rating prediction is obtained. The experimental results show that this approach can improve the recommended quality.     

Key words: collaborative filtering;cloud model;user characteristic attribute similarity;scoring preference;cloud similarity