Studying the Performance of QoS Specific Web Service Recommendation System Using Virtural Regions

The emergence of Internet and web services made a tremendous impact on the data retrieval of the users. The users are sophisticated to utilize the web services based on the recommendation systems. There are various categories of recommendation system reported in the literature. Web Service Recommender Systems (WSRS) based on Collaborative Filtering (CF) achieves best Quality of Service (QoS) results. The users are classified based on the similar IP addresses and regions are created. In this paper a new recommendation system is proposed where the users are classified and grouped together based on virtual regions. The virtual regions are found based on the QoS parameters computed based on round trip time of the services. The proposed Virtual Region based Filtering algorithm (VRF) for web service recommendation significantly improves the prediction accuracy and time complexity. The developed approach is tested with real-world web service QoS data sets. The proposed system achieves improved performance with respect to the parameters such as Round Trip Time (RTT) and Mean Absolute Error (MAE).