The Designment And Implementation In Information Retrieval For WEB Mining Based On Ontology

In the respect of information retrieval, users' need can't fulfilled by the the traditional technology of information retrieval based on the key words and subject classification increasingly, Furthermore, the difference of network user's knowledge background and Natural Language Ambiguity lead to the lower recall and precision of the online information retrieval.Because of the good concept Hierarchical structure and the support to logical reasoning. Ontology has been widespread concerned in the information retrieval especially in the retrieval based on the knowledge, which is also the hot spot of the information retrieval.In response to the shortcomings and deficiencies exist in the information retrieval methods, the thesis applies the Ontology technology to Intelligent Information Retrieval.This thesis makes the study and practice on the theory methods, techniques and tools of the Ontology, Applying book online system. The thesis analyzes Ontology, domain Ontology and Semantic Web studies the relevance of Ontology technique in the WEB Application Subject Classification, and then with the help of the protege and the combinination of Ontology technique, the theies proposes a new method of intelligent retrieval and a model of intelligent search based on Ontology. The final part of the thesis carries the theory into execution which establishing a domain-based intelligent search engine model based on Ontology technique, followed with a verification of experiment.The major contributions of the thesis are as follows:(1) Firstly, the function of book intelligent retrieval system is analyzed , then pointing to its framework model and detailed function module.(2) W3C standard OWL is choosed as the Ontology description language, implementing the combination of systems and ontologies and then constructing a Ontology library of the fields of computer professionals.(3) Based on the Ontology library that has been constructed, RDF/RDFS is as the semantic labeling languge for the original data of book informations, and Jena is as the main tool, Extended reasoning and data query are realized in this system.