1.School of Electronics Engineering and Computer Science, Peking University, Beijing 100871, China; 
2.Key Laboratory of High Confidence Software Technologies, Ministry of Education, Peking University, Beijing 100871, China

In the recent years, the web has kept growing rapidly and undergone tremendous changes towards a user-centric environment. With the proliferation of services available on the Internet, millions of users are able to voluntarily participate in and collaborate for their own interests and benefits by means of service composition. However, due to the ever increasing number of services, it then becomes a challenging issue to enable the users to rapidly select and compose the proper services. In this paper, we propose an approach which utilizes service communities for automated service composition. This paper makes three contributions. Firstly, we propose the service community architecture and identify two key components to facilitate service discovery and composition. Secondly, assisted by the service community, we provide two composition types to help users explore the space of potential composition opportunities without having to understand too many details of individual candidate services. Finally, we evaluate our approach by performance analysis and provide a web-based prototype that lowers the entry barrier for the users.

This work was supported by the National Basic Research Program of China (Grant No. 2009CB320703), the National Natural Science Foundation of China (Grant Nos. 60821003, 60873060), and the High-Tech Research & Development Program of China (Grant No. 2009AA01Z16).