Design of a Recommendation Filtering System in Mobile Commerce
Zheng Wan · 2009
This mobile commerce has become a research hotspot in recent years for its commercial value and technical maturity. While more and more service providers participating in mobile commerce promote its popularity, problem called "spam recommendation" decreases users' satisfaction. Therefore, efficient mechanisms should be adopted to improve quality of recommendation. In this paper, a recommendation filtering system is addressed to reduce arbitrary recommendations. The general idea is to record users' operations (called "implicit feedbacks") on recommendation with the form of short message to obtain users' interests and to update user profiles continuously. At the same time when receiving new recommendation, the system computes the similarity between user profile and recommendation message, and then judges whether the message is a spam.