Enhanced privacy preserved web service recommendation with clustering

As the number of web services are increasing day by day, it is difficult for users to choose the best service from it. Here we can see the importance of Quality of Service (QoS) based web service recommendation that helps us to select the best service. The existing systems are using the collaborative filtering techniques for the personalized QoS prediction. The recommender systems collect QoS values from users and this may affect the user's privacy. Due to this reason, users do not want to reveal their data to the service provider which will badly affect the performance of recommender system. Since privacy is an important thing in this growing world, this paper considers privacy for recommender systems. A data masking technique i.e. data obfuscation is used for privacy protection. To improve the process clustering is incorporated. In order to recommend for services, its functional and non-functional similarities will be measured. After that process, finally the top K recommendation list will be produced.