| With the explosive growth of Internet, World Wide Web (WWW) has become an enormous, dynamic information service network with multiple information resources and worldwide sites. Traditional information retrieval technologies can not meet the need for better understanding and processing information on the semantic level. However, based on semantic understanding, semantic retrieval has become a hot topic. Because of ontology's capability to represent meaning of the information by a hierarchical structure, and its reasoning support, it is significant to use ontology in information retrieval.Ontology, as the basic of Web information sharing and exchanging on the semantic level, plays a key role in semantic retrieval. Providing an ontology which contains all knowledge is impossible. At present, most researches of semantic retrieval-based ontology assume ontology contains all knowledge, and have not considered the impact on having no available domain ontology on information retrieval performance. In addition, they have not made full use of the relations of concepts in domain ontology.To solve this problem, this paper proposes a hybrid information retrieval model based on domain ontology. It combines keywords and semantic retrieval. Even if there is no available domain ontology, it can work based on keywords. Meanwhile, to efficiently mining semantic information, it makes full use of semantic relevance in domain concepts.In addition, in order to test the performance of the hybrid information retrieval model, this paper constructs the domain ontology of computer network, and experiments by comparing hybrid model with traditional keywords retrieval model. Experiments show that the hybrid model is better than traditional model.The development of semantic retrieval based on the domain ontology depends on AI and NLP's development. The study of this paper is just a beginning step of the research in this field. There exist a lot of problems in the model and a further study is given. |
PDF Full Text Request