Research On Ontology-based Semantic Retrieval

Information Retrieval is a procedure that extracts related information and documents from the data sets. Along with the increasing accumulation of data and the rising demand of high quality of retrieval results, it has been very difficult for the traditional IR technologies to meet task of high-quality information retrieval. As a newly emerged knowledge organization system with high starting point, Ontology technologies are playing a very important role in promoting the function of IR in knowledge management.At present, Most of the information retrieval systems are based on mechanical retrieval technique in the syntax layer. The system may miss some information that users really want and return some information that users don't want, after the investigation on the concerned theory of IR technologies and the knowledge of Ontology technologies, in order to overcome the shortcomings of traditional vector space model in dealing with semantic, we conducts an IR system construct model based on fuzzy Ontology. With the help of the domain knowledge in Domain fuzzy Ontology, it explains how to realize concept-based retrieval. In this model, concept mapping and similarity calculating are made based on the membership-degree concept of fuzzy ontology, which make it possible for us to analyze document on concept level.The system model contains two parts: document pretreatment and document retrieval. In the IR model, it adds an Ontology server, which is different from the traditional IR system construct. With the help of Ontology, the document pretreatment transacts and indexes the retrieval sources based on Ontology,, the query conversion module does some semantic processing on user needs and expanses the initial query on its synonym,hyponym and its senses; the retrieval agent module uses the conversion of queries to retrieval the information sources. In the specific realization process, it makes an in-deep and pinching the results using the algorithm A priori.Finally, with the help of researching results we present a novel evidence analysis method in computer forensic domain based on Ontology. Experiments are carried out to test it performance, and the result shows its efficiency.