Research And Implementation Of The Telecom Package Recommender System Based On Consumer Behavior

With the rapid popularization and development of the fourth generation mobile communication technology, more and more mobile phone users choose to use 4G telecom packages. In the 4G era, the mobile data traffic increase rapidly, users' requirements and preferences are changing dramatically, showing a personalized trend. As the same time, telecom operators develop and launch a lot of new 4G telecom packages to meet market demand. It becomes difficult for users to select suitable products for themselves because of too many choices, which also brings challenges to operators' management and marketing. Therefore, operators are eager to find out the users' requirements and preferences from a huge amount of users' data, and provide efficient, accurate and personalized recommendation for each user.The current package recommender methods have the problem of low efficiency and precision, in this paper, we research a hybrid recommender algorithm and build the telecom package recommender system based on consumer behavior, combing data mining, collaborative filtering and utility-based recommender algorithms. With the full investigation and study of the related technology, firstly we research the pricing strategy of the telecom packages and give some suggestions for the package design. Then, we study users' consumption data to find out users' preferences and potential requirements. Based on the user's consumer behavior model and telecom package information, a hybrid recommender algorithm is designed, and we compare the precision with other two algorithms with the actual data of the operator. Finally, we design and build the recommender system based on consumer behavior by applying the hybrid recommender algorithm.