Study Of The Information Retrieval System Model Based On Ontology

With the rapid development of Web, the information on the web is becoming rich more and more, and it has been the largest information database that human beings have ever had. People depend on various kinds of information retrieval technologies to locate information that they want, but the explosion and variety of information bring different difficulties to the information retrieval. The current information retrieval tools provide searching accord to keyword merely, and neglect the semantic meaning contained by keyword, so the recall and precision are low and return too much results.Based on the existing information retrieval technologies, we make in-depth studies on the Ontology-based information retrieval, and propose a novel information retrieval model, then we analyze and study the main problems in query building, documents processing and query matching in the model. The main tasks are as follows:1. Make in-depth studies on current information retrieval technologies,Existing problems and reasons of the problems;2. Introduce the basic knowledge about Ontology, discuss the application of Ontology in information retrieval, propose a Ontology-based information retrieval model, and design the functional modules of system detailed;3. Introduce a new knowledge representation method, conceptual graph, and compare it with other methods. After the defining of concepts similarity, relations similarity and conceptual graphs similarity, we propose a algorithm to search information by matching of conceptual graphs. It improves the efficiency of information retrieval dramatically;4. Study the building method, building principle, and building process of Domain Ontology, and build a simple Domain Ontology of computer science.