Electronic Official Document Inquiry System Based On Ontology And Lucene

E-office is a system solution adopted by the government to shift government functions such as information publication, management, service and communication to the internet, aiming at increasing government working efficiency, raising transparency, improve decision-making and investment environment, reinforcing effective management of the economy and society and lifting legalization level.As the promotion of E-office, more and more electronic documents are produced by various departments. It's is increasingly important to effectively manage these electronic documents and provide efficient index mechanism in order to make sure that the user can find whatever materials they are interested whenever they want. In the E-office, most departments adopt file management system that indexes the files by matching the key words. The defect of this indexing method is that it fails to understand the semantics of the key words input by the users. So understanding the semantics of the key words input by the users will help increase the complete rate and correct rate of the inquiry, so as to satisfy the user's inquiry needs more efficiently.In dealing with the above-mentioned problems, a pilot index based on the traditional document index methods is addressed in this article. This method can inference based on the inquiry key words input by the users, then it will list some relevant inquiry suggestions for the users to choose. In this way, it can increase the complete rate and correct rate of the inquiry, so as to increase its average performance.The research adopts the ontology development tool Protege3.2Beta of Stanford University to express the electronic documents in e-office domain, OWL-DL as the description language of ontology, JESS (Java Expert System Shell) as the inference engine. At the same time, it adopts JESSTab to fulfill the conjunction between Protege and JESS, while uses Lucene as the index engine core for full-test index. As the research is based on the documents publicized by various departments of Jiangxi University of Finance & Economics, the full-text search provided in the system is directed at word format documents. At the same time the search is directed at Chinese information. So it is necessary to extract the context of the word format documents before performing full-text research on the information. POI tool is adopted in this research to extract context of the word format documents. In addition, as the search is directed at the Chinese word format documents, word splitter operation should be performed on the Chinese context extracted after extracting the context from word...