Web Service-oriented Data Mining Framework Model For Extending The Accidental Services

Web Services set is one of the best ways to provide resources and services, which is widely accepted by company, individual, intermediate service providers and other service providers. Currently the web is going through a transformation from a data-centric Web to a Semantic Web consisting of both self-describable data and Web Services. This trend is providing us with increasing opportunities to find and compose potentially interesting and useful services from the increasingly rich service compositions. However, with the rapid development of today's Web Services, on the internet, there is a set of large, chaotic scattering set of Web Services. how to find the services that we need or the customers need, is difficult to achieve efficiently.This paper presents a framework for Web services data mining model. In the model, all objects that we used are in the form of a Web services object, all of the tools and functions that we used, including data mining tools, information processing tools, are also in the form of a Web service object, all of the input and output data are in the form of XML, which can be limited and regulated by DTD, and the service interfaces of the whole framework that provide services to others still is in the form of Web services object. As the semi-structured nature that XML itself has, makes data import and export without going through the complex process, which has avoid the difficulty of heterogeneous data resulting from the application. In the framework, we deal with in the implementation process through the implementation of Web services approach, which makes the entire framework is no longer limited the tools should come from itself, the tools we need can be placed on any other server on the internet. Therefore, the framework model can be easy and convenient to use in various areas.In this paper, we take the actual test work application for the concrete realization of the overall model– extend the unexpected service during information search in the Web service. In the actual testing project, we focus on the data mining on the web services, and composing potential interesting, useful services more efficiently through the evaluation of the semantic information of Web services, especially the unexpected but useful services, which we call the Unexpected Services. And for other areas, we can call the similar results of accident as the unexpected results set. The extension of the unexpected results set, is valuable in many areas. In the searching service, unexpected results can be a rich set of results matching sets, provide the result intelligently for users than the direct search which can be more accurate and useful; in research, like medicine, geography, and other areas of expertise, unexpected results can be easily extended more possible results during the accurate processing; in the network business, the unexpected results can be seemed as an extension of the excavation process for the user needs to get information that what the user may be interested, and what the customer may want to buy.In this paper, we try to use some of the special extensions to extend existing Web service ontology during the transformations of WSDL to OWL-S. We try to identify the types of the relationship of Web services on a general level. We try to use the Observer Pattern of Design patterns to transform top-down searches into bottom-up matches, which provide the ability of composing unexpected and useful services for the framework, and the evaluation of the interestingness and usefulness of the screened service compositions. So that we can do further filtering and choice to make the service composition results more accurate.