Domain Ontology-based Information Retrieval Research

Information Retrieval is a procedure that extracts relatedinformation and documents from the data sets. Along with the increasingaccumulation of data and the rising demand of high quality of retrievalresults, it has been very difficult for the traditional IR technologies tomeet task of high-quality information retrieval. As a newly emergedknowledge organization system with high starting point, Ontologytechnologies are playing a very important role in promoting the functionof IR in knowledge management.At present, most IR systems are just based on keyword-matchingfor the needs of users. In this thesis, after the investigation on theconcerned theory of IR technologies and the knowledge of Ontologytechnologies, it conducts an IR system construct model based onDomain Ontology. With the help of the domain knowledge in DomainOntology, it explains how to realize concept-based retrieval.The system model contains two parts: document processor anddocument retrieval. In the IR model, it adds an Ontology server, which isdifferent from the traditional IR system construct. With the help ofOntology, The document processor tags and indexes the retrieval sourcesbased on Ontology; the query conversion module does some semanticprocessing on user needs and expanses the initial query on its synonym,hypernym,hyponym and its senses; the retrieval agent module uses theconversion of queries to retrieval the information sources. In the specificrealization process, it makes an in-deep study of the key part in retrievalprocess—query expansion and pinching the results using the algorithmApriori, and at last conducts an information retrieval system based onWordNet.