Research On Semantic Data Mining Technique Of Chinese Information

Due to the lack of a unified semantic description, there are plenty of complex and duplicate information in traditional information. Facing with massive information, how to obtain useful information quickly and effectively from the "Information Ocean" is a very difficult problem. Introducing semantic information into computer information process is a fundamental solution to this problem, and can achieve better sharing of information.Because of the special nature and complexity of Chinese information, it's difficult to process Chinese information relatively. Semantic information put a higher demand on Chinese information processing. Users no longer meet the direct access to information only, and need to get more implied semantic information. data mining comes out for this. However, traditional data mining need experts's help in the field and rely on data-driven. Gradually it's unable to meet the needs of users. Ontology is the formal description of the objective knowledge. Data mining combined with semantic analysis technology based on ontology can solve information semantic processing problems and contribute to realization of Chinese semantic data mining.Supported by MII's electronics industry development fund, the semantic data mining technology of Chinese information is studied in this thesis. Firstly, aiming at the characteristics and key technologies of Chinese information processing, especially Chinese word segmentation techniques, a Chinese word segmentation algorithm of Max Matching and Dictionary is designed. Secondly, Ontology-related knowledge is introduced, and the Semantic data mining technology based on Ontology (OSDM) is proposed. Its workflow and principles are given, and its key technologies are analyzed, including Ontology construction, semantic annotation and semantic reasoning, etc. Practical solutions are also described in detail. This lay a good foundation for semantic data mining technology applications. Finally, based on OSDM model, a Chinese Semantic Information Retrieval System is developed. In this system, a MyFruitOnto field ontology is constructed as knowledge base, and semantic reasoning ability of OWL ontology language is used to expand the key words reasoning for obtaining more accurate user intent and returning more correct results. Meanwhile, in order to overcome the limitations of the field, the system also has full-text search function for providing a better user experience and more fully functional and more reliable system performance.Initially it realizes the Intelligent Information Retrieval.