Service Recommendation with Context-Aware User Reputation Evaluation

Abstract

There is a growing number of services in the age of information. Therefore, choosing a satisfactory service has become increasingly difficult. One way to solve this problem is to construct a recommendation system. Selecting a set of users called neighbor user set and then generating the recommended services through collaborative filtering is a commonly used method to implement a recommendation system. In the recommended process, the user reputation is always considered since we prone to trust the users with high reputation. However, the current recommendation System lacks detailed context information analysis in user reputation calculation. In this paper, we propose a context-aware reputation calculation method based on user ratings. We first propose a novel method to analysis the impact of different contexts on user ratings. Then, we apply the context information of target user to obtain neighbor user set. Finally, we generate services for target user based on his neighbor user set. Experiments show that our method has better accuracy than other traditional methods.