With the development of social informationization, large amounts of information ofgovernment departments is stored in the form of electronic data, and the level of the electronictransaction processing is higher than before. But the main form of text processing is still inthe form of documents, which makes all kinds of text data has not been more widely used.This thesis make in-depth analysis of the feature of information distribution of text inaudit-field of social insurance based on the audit of national insurance, and research thetechnology of processing the text in audit-field of social insurance aimed to the application oftext of audit retrieval.This thesis first discusses the development of text information retrieval, the basicprinciple of information retrieval and the research situation of information organization andretrieval. Second, for the features of unstructurization of text, this thesis discuses theorganization of text in audit-field of social insurance, combining with pre-processingtechnology of information retrieval and ontology. Then, this thesis discuss the applications ofquery expansion in information retrieval, shows semantic expansion strategy based on theaudit ontology. Finally, this thesis shows the effect of the performance of the informationretrieval system, and verify the actual effect of query optimization proposed in this thesis withthe social pension insurance relevant text data. As a conclusion, Information retrieval systemof text of audit based on query optimization is applicability in this field. |