Abstract
User reliability is notably crucial for personalized cloud services. In cloud computing environments, large amounts of cloud services are provided for users. With the exponential increase in number of cloud services, it is difficult for users to select the appropriate services from equivalent or similar candidate services. The quality-of-service (QoS) has become an important criterion for selection, and the users can conduct personalized selection according to the observed QoS data of other users; however, it is difficult to ensure that the users are reliable. Actually, unreliable users may provide unreliable QoS data and have negative effects on the personalized cloud service selection. Therefore, how to determine reliable QoS data for personalized cloud service selection remains a significant problem. To measure the reliability for each user, we present a cloud service selection framework based on user reputation and propose a new user reputation calculation approach, which is named MeURep and includes L1-MeURep and L2-MeURep. Experiments are conducted, and the results confirm that MeURep has higher efficiency than previously proposed approaches.
Highlights
In the age of Internet of Things (IoT), cloud services have been the widespread concern in many realms [1,2,3]
The reputation calculation model named MeURep is represented to calculate the reputation of each user based on the historical quality of service (QoS) data provided by the users in personalized cloud services
Su et al [21] studied the trust perception approach for service recommendation and used QoS values to calculate the user reputation based on clustering algorithms
Summary
In the age of Internet of Things (IoT), cloud services have been the widespread concern in many realms [1,2,3]. The users with high reputation will provide reliable QoS data, which generate more reliable conditions for other users to invoke suitable services. According to the unreliable QoS data, users may select unsuitable or bad services. To address this issue, it is necessary to evaluate the reliability of users in the cloud services environment. We present a user reputation calculation model, which is named MeURep. In our approach, the reputation calculation model is based on the historical QoS data submitted by other users. 2. The reputation calculation model named MeURep is represented to calculate the reputation of each user based on the historical QoS data provided by the users in personalized cloud services.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have