A Method Service Selection Based On The Hybrid QOS

With the rapid development of Internet, Web service is widely used, at the same time,the related technology has a great progress and development. A large number of Web services has emerged on the Internet, in order to select the most appropriate service, firstly,the service requester usually search the service which can meet their functional requirements, but multiple services with similar functions published on the Internet can satisfy the requests’ functional requirements, so user needs to select the services based on the non-functional properties of services,especially the Quality of service QOS (Quality of service), thus from the perspective of QOS attributes, this article focuses on the service selection method based on hybrid QOS.From the public literature, we can find the current methods of service selection that they all assume all QOS attribute values of services are determined in real numbers, it does not take into account the fuzziness of QOS properties.However, this assumption can’t reflect the reality,therefore, this article gives the relative superiority degree algorithm base on hybrid QOS attributes for service selection, and then through the simulation and experimental validation, illustrates that the algorithm of relative superiority degree is very validity and rationality.This paper mainly do working as following:1. Introducing the Web service theory and the related technology, including:the architecture of Web services, the core technology of Web services and the choice of service technology etc.2. Discussing the QOS properties of Web service, it firstly presents the definition of QOS, and classify the QOS attributes. Considering the QOS attributes are fuzzy, it describes the QOS attribute value on the basis of precise numerical value, interval number and triangular fuzzy, and analyzes the advantages of the description ways. At the same time, it describes users preferences’ demands for different QOS attributes with the order relation vector, and convert them to the subjective weights of QOS attributes, then the objective weights of QOS attributes are obtained by using the entropy weights.3. Putting forward a service selection method supporting of mixed QOS attributes, it firstly introduces TOPSIS (Technique for order preference by similarity to an ideal solution) algorithm, then analysis the algorithm is insufficient, so the algorithm of Relative advantage degree is proposed. Finally it gives the algorithm of relative advantage degree based on hybrid service QOS attributes.4. Through simulation and experiment, it compares the relative advantage degree algorithm with the TOPSIS; And it proves the algorithm of relative superiority degree has high sensitivity to observe the change of QOS attribute values.Then it compares the pattern of combing subjective with objective, subjective and objective for selection of Web services, it proves the pattern of combing subjective with objective is rationality; at last it shows that the model is also suitable for multi-users service selection, it can reflect the needs of different users preferences.