Itembased collaborative filtering method using Pearson Correlation Coefficient

Cloud computing is becoming popular. A number of works have been carried out on cloud computing , including performance analysis, market-oriented cloud computing, management tool, workload balance, dynamic selection, etc. Quality-of-service has been widely employed for presenting the nonfunctional characteristics of the software systems and services . QoS of cloud services can be measured from either the client side (e.g., response time, throughput, etc.) or at the server side (e.g., price, availability, etc.). Based on the service QoS measures, various approaches have been proposed for service selection , which enables optimal service to be identified from a set of functionally similar or equivalent candidates. To provide QoS ranking information for the service selection approaches, this paper focuses on predicting QoS ranking of cloud services. Collaborative filtering methods are widely adopted in recommender systems . A memory-based approach is one type of the most widely studied collaborative filtering approaches. The most analyzed examples of memory-based collaborative filtering include user-based approaches , item-based approaches. User-based and item-based approaches often use the vector similarity method and the PCC method as the similarity computation methods. Compared with vector similarity, PCC considers the differences in the user rating style when calculating the similarity. The rating-based collaborative filtering approaches try to predict the missing QoS values in the user-item matrix as accurately as possible. In the ranking-oriented scenarios, accurate missing value prediction may not lead to accuracy ranking prediction.