A Personalized Service Recommendation Algorithm for Service Functionality

With the increasing number of services on the Internet, it has become a great challenge to help users find services according to their demands. Personalized recommendation technology is an effective way to solve the problem. Existing service recommendation approaches make recommendations among services with same or similar functionalities to meet the non-functional requirements of users, while the functional requirements for service are seldom taken into account and new services that satisfy the needs of users are difficult to be recommended. Therefore, in this paper, we introduce social tags to the process of service recommendation and build service functionality oriented social tags model to describe user preference for service functionality, then we present a personalized service recommendation algorithm for service functionality (PSR-SF). The proposed algorithm first discovers the neighbors of a target user according to services use frequency of users, and then clusters services that have been used by the target user and his neighborhoods by using the functional characteristic vector of services which based on service functionality oriented social tags model. Finally, target user preference for service classes are generated by using service functionality tags use information of users to make recommendations. The experiment results show that the performance of PSR-SF algorithm is better than those existing recommendation algorithms in terms of service recommendation precision, recall and F values.