Research On Semantic Web Mining

Research efforts on semantic web are meant to enhance the current Web with meta data and to make the web data rich with meaning in order that computer can understand and process web data automatically. Web mining resorts to data mining techniques to extract and discover useful information from the ocean of web documents and services. The idea of semantic web mining is to exploit the newly-emerged semantic web technique to improve the web mining. On the other hand, web mining can also be very helpful for the creation of semantic web.We focus our discussion on the research of semantic web mining in this thesis. The major work can be summarized as follows:We firstly introduced the related necessary theories of semantic web and web mining, and explored the way of combining them together. Then, after the analysis of the semantic hierarchy of RDF (Resource Description Framework) resource description, we proposed a kind of RDFMS (Resource Description Framework Model and Syntax) data hierarchy clustering method based on semantic distances among data. We came up with algorithm descriptions and reinforced our conclusion by means of real testing examples.Some data mining techniques, which can be applied to semantic web mining, were also discussed in this thesis. We explored the suitability of Inductive Logic Programming (ILP) method in semantic web mining in more detail, and showed how to make use of this method in the semantic context.Based on all the work discussed above, we concluded our thesis by putting forward one semantic web knowledge discovery model with the help of agent technique, and finally, illustrated the functions of all subsystems.