Call Data Record Based Recommender Systems for Mobile Subscribers

Abstract

Recommendation systems for mobile phones are of great importance for mobile operators to achieve their desired profit targets. In a client inferred market, the number of contract users and contract phones is especially significant for mobile service operators. The tremendous growth in the number of available mobile cellular telephone contracts necessitates the need for a recommender system to assist users discover suitable contracts based on their usage patterns. This study used a hybrid of both collaborative and content-based filtering. A prototype of a mobile recommender system was developed and evaluated using precision and recall. The developed recommender system was able to successfully recommend packages to subscribers. A precision-recall curve was produced, and it showed good performance of the system. This study successfully showed that a hybrid system was able to recommend products to the mobile subscribers.