Intelligent Information Retrieval Based On Semantic Web

With the rapid development of Internet technology and the increasing of information resources, the demand of getting information resources is also continuing to enhance. It become more important on how to get desired results quickly and accurately from a large number of information resources. Traditional information retrieval mainly provided the user interfaces which is based on keyword search. In most cases, the user can not get the satisfactory results, because it is difficult to express the real intention of the user search for a combination of a few keywords. The main objective of the subject is identify effective calculation of the similarity of the Semantic Web based on semantic web technology, make information retrieval based on the semantic level improving from key words and complete the intelligent information retrieval experiment system MobileSearch based Semantic Web.This paper researches on the calculation of similarity of the Semantic Web, the similarity algorithm of Semantic Web is given out. The main idea is that turn the semantic web of users query and the semantic web of raw information into a directed graph, achieve calculation of similarity Semantic Web by matching directed graph. The algorithm achieve semantic matching and improve recall and precision of information retrieval.This paper build the Semantic Web by extracting and analyzing the original information resources, and store it to the Semantic Web library. When the user retrieval, first make search criteria into semantic web of user conditions, then match the semantic web of user conditions and the semantic web of original information by the similarity algorithm of semantic web given out, then search results will be returned by similarity.This paper designed and implemented information retrieval intelligent experimental system MobileSearch based on semantic web. In this system, build a Semantic Web library of mobile phones. MobileSearch provide a good user interface, can provide users semantics guidance, then achieve semantic retrieval. The system is for a certain amount of experimental, analyzing the experimental results, recall and precision are greatly improved by comparing with traditional information retrieval.