Utilizing Customer Satisfaction For Ranking-oriented Cloud Service Recommendation

With the rapid development of the Internet,cloud computing technology has received widespread attention from academics and industry.Cloud services derived from the cloud computing environment have also been popularized and promoted.Cloud services are virtual commodities deployed in the cloud.A large number of cloud services with the same functional properties but different non-functional properties often make it difficult for users to choose.Among these,user satisfaction plays a decisive role in the survival of cloud services.Therefore,how to find the best and the most satisfying services for users has become a problem to be solved in today's cloud service field.The cloud service recommendation algorithm mines and predicts the user's personalized preferences through the user's behavioral characteristics,and forms a recommendation list for the user in accordance with his preferences.The cloud service recommendation system generated thereby is the key to solving the above problems.In this paper,we consider the impact of similar users and their own attitudes and expectations for cloud service selection,we propose a cloud service ranking recommendation method that considers user satisfaction.This method will help discover the cloud service with the highest user satisfaction and recommend the service to the corresponding user.This paper includes the following two main tasks:(1)When calculating user similarity,an enhanced ranking similarity measure method based on Jaccard factor is proposed.This method can measure the similarity between users more accurately.After experiments,the recommendation results produced by this similarity measure method are not only high accuracy in terms of numerical value but good performance in sorting results;(2)In the forecasting process of cloud services,a user satisfaction measurement method was proposed.This method can simulate the user's attitudes and expectations in the selection of cloud services.It also designs a cloud service ranking recommendation process that combines user similarity and satisfaction to obtain the more accurate and higher user satisfaction recommendation results,the experimental results show that the method proposed in this paper,with strong applicability and effectiveness,has higher prediction accuracy than the existing methods,and can satisfy various user preferences through parameter adjustment.