A Web Services Composition Discovery Approach Based on Service Data Network

With the rapid development of service-oriented computing, an overwhelming number of Web services has been published online. Even though existing Web services possess rich functions, it is still challenging to satisfy individual and personalized requirements of different users. An ideal way to overcome the problem is to compose existing Web services together and construct a Web service composition which can implement complicated and characteristic functions. To this end, we focus on the problem of constructing Web service composition based on service data network in this paper. Due to the inner relationship between services (e.g., data communication, data interdependent), service data network is constructed based on services' input and output information. Additionally, we propose a Web service composition discovery method based on the service data network, which fulfills user requirements. Experimental results on real-world dataset demonstrate the effectiveness of the proposed approach.