A Service Ranking Method Based on Collaborative Filtering

Collaborative filtering is one of the most popular service recommendation techniques. When more and more web services are published online, this technique can recommend personalized services to users, which more satisfy user´s preference. In this paper, we presents a service ranking method which based on collaborative filtering, it contains two parts: neighbour user selecting and service ranking. In neighbour user selecting, we provide a mixed neighbour user selecting method containing user service requirement similarity calculating and user rating similarity calculating. In order to avoid cold start problem, the final ranking score is a combination of user similarity and service QoS requirement.