Recommendation method for mobile network based on user characteristics and user trust relationship
Peng Fei Yu · 2016
By analyzing the common sparse problems of the score data in the collaborative filtering recommendation system, this paper presents a recommendation method for mobile networks based on user characteristics and user trust relationship. In this method, the similarity of user characteristic attributes is used to modify the similarity calculating of the original score. Then, according to the communication data of mobile users, a trust value calculation method is put forward to meet the concept of social science. Finally, by the adaptive coordination factors to process the similarity measure results of the two aspects, the final project prediction score is obtained. Experiments show that the proposed method can effectively alleviate the problems caused by the sparse user score data, and improve the prediction accuracy of mobile recommendation system.