Study on the Practice of Ontology in Information Retrieval System
|School||Liaoning Normal University|
|Keywords||ontology semantic indexing user interest model intelligent information retrieval information retrieval model|
With the rapid increment of the information on the web,Traditional information retrieval based on the keywords is far from user’s satisfaction in recall and precision. As a new way of knowledge description,ontology will have a wide prospect of application in search engine area because of its capability to represent meaning of the information by a hierarchical structure and its reasoning support.On the one hand,The information retrieval can be promoted from the keywords level to the conception level by ontology,It will promote the recall and the precision;On the other hand,The relevance of the conception can be exploited and reasoned,It will realize intelligent information retrieval. This thesis gives in-depth analysis and discussion about the function of the ontology in three aspects of information retrieval,including the analysis and construction of user’s demand、The analysis and semantic indexing of targeted documents and the construction of user interest model,And then an intelligent information retrieval model based on ontology is given.The main contributions are as follows:1.Propose an ontology-based semantic indexing method. The method takes advantage of ontology to clear the conceptual ambiguity and then build the reflective relationship between ontology and the web document,which realizes the semantic indexing of targeted document.2.Propose an ontology-based intelligent query method. The method takes full use of the function of ontology reasoning,and gives a full consideration of concepts、attribute、relationship and other information in users’query,which can find out what users really need.3.Propose and implement an ontology-based intelligent information retrieval model. The realization of this model not only will promote the recall and the precision but also makes the retrieval system more intelligent and personalized.