| In recent years, with the development of e-commerce, Semantic Web, XML technology and all the other internet technologies, the internet environment become highly complex and the user have developed higher expectation on the functionality and experience of internet product. This made the original UDDI(Universal Description, Discovery and Integration)based or Semantic based Web service discovery technic fail to fulfil the demand of modern user.For this reason,to find the web service best fit user demand, in a poll of web service with similar function and structure, have become one of the hot topic in the field of internet technology.With studying current Semantic Web,UDDI(Universal Description,Discovery and Integration) based and Semantic based Web service discovery, OWL-S protocol, this paper covers topics listed below.First, A model based on mix and match UDDI and semantic similarity is developed,followly, the function and reaction between each module is analyzed. Use hybrid scheme to increase speed of match and decrease response time, achieve user experience improvement. Second, we propose a dynamic metrics algorithm that calculate semantic similarity of services as well as the method taht calculate dynamic metrics is discussed.Introduce FCM based service cluster pre-process to increase match speed. The semantic-distance -property similarity based algorithm is improved by unbalanced ontology tree-way and extend-way. This algorithm quantifies service quality QoS, which is not very well supported by OWL-S. This paper present a way to quantify specific QoS, as well as normalization method.Experience have been conducted on the proposed improved Web service discover model, the result show the model can be efficiently executed. Meanwhile, the experience data analysis proves the practicability, superiority and value of the UDDI and semantic similarity mix and match based model. The limitation of the model and future works is also discussed in this paper. |
PDF Full Text Request