Collaborative Filtering Service Recommendation Algorithm Based on Trusted User and Recommendation Evaluation

摘要

Collaborative Filtering (CF) is a widely used service recommendation technology. However, most service recommendation algorithms based on collaborative filtering do not take into account the influence of some abnormal data and malicious evaluation data in user feedback data on the recommendation accuracy. For the trusted requirements of service recommendation in cloud computing environment, this paper proposes a service recommendation model based on users services network (USN). The model calculates the user's recommendation to the service and the user's trust by establishing a recommendation relationship between user services and the trust relationship between users, and by considering the influence of factors such as malicious users on recommendation and trust, increased authenticity and availability of recommendation and trust. Based on this model, a collaborative filtering service recommendation algorithm based on trusted users and recommendation evaluation is proposed. Experimental results show that the algorithm is superior to other mainstream service recommendation algorithms in improving recommendation accuracy and regulating malicious evaluation behavior.