Comprehensive Ranking For Web Services Throhgh Considering And Merging Multi-Source Information

In recent years, with the constant development of network technologies and web applications. Web service as a platform-independent and programmable web application featured in openness, loose coupling and self-describing has attracted extensive concern of numerous scholars at home and abroad on academia and industry which makes it an issue in research in recent years. With the constant development of web technologies and the increasingly released number on web service, there appear a large number of web services with the same or similar functions. As a result, people have more choices that promote their high requirements on the quality of service. How to provide such a web service that both can satisfy functional user requirements but has the Qo S guarantee now has become the focus of people’s attention. Along with the growing number of this kind of web service, its selection has become a hot spot in the field of service computing.For the past few years, more and more researches have been done on web service selection, among which has become a focus and more work is putting on the ranking algorithm which is based on Qo S. This paper presents two comprehensive ranking of web service algorithms for prediction on web service ranking. One is called TOPSISRank that is based on a multidimensional Qo S and another one is called Qo SRandomRank that is based on random walk. TOPSISRank algorithm applies the idea of TOPSIS mode and puts comprehensive consideration on a number of different Qo S attributes on service to sort projections for candidate services. The algorithm will transform Qo S-based web service selection problem into a multi-objective decision problem and use the common-used TOPSIS decision-making method of the information entropy. Finally, results of web service sequence prediction are obtained. In addition, in order to verify the validity of the TOPSISRank algorithm of web sequence forecasting, this paper has done an experiment on TOPSISRank by using real data sets QWS and made a contrast with another prediction method of sorting the WSRF algorithm. The experimental results show that the TOPSISRank prediction in the Qo S-based web service has achieved terrific effect on prediction. Qo SRandomRank algorithm applies the idea of random walk mode and takes through the Qo S attribute value to determine the superior relations between the services whose probability is applied in sorting projections for service. Through the research of this article, the proposed sorting algorithm has provided a certain reference on such a kind of web service that can satisfy the needs of its own web service for users to select the most.