QoS prediction for Web service in Mobile Internet environment

QoS prediction plays an important role in Web service recommendation. Many existing Web service QoS prediction approaches are highly accurate and useful in Internet environments. However, the QoS data of Web service in mobile Internet is notably more volatile, which makes these approaches fail in making accurate QoS predictions of Web services. In this paper, by weakening the volatility of QoS data, we propose an accurate Web service QoS prediction approach based on the collaborative filtering algorithm. This approach contains three processes, i.e., QoS preprocessing, user similarity computing, and QoS predicting. We have implemented our proposed approach with an experiment based on real-world and synthetic datasets. The results demonstrate that our approach outperforms other approaches in Mobile Internet. Keywords-Web services; QoS; collaborative filtering; similarity computing