Research On Ontology-Based Semantic Information Retrieval

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 needs for better organizations, understanding and processing services in the semantic level; therefore, to find new ways has also become a hotspot of current research.Based on the existing semantic retrieval methods, 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 user queries, documents processing, semantic retrieval and results optimization in the model. The main contributions are as follows:1. Propose an ontology-based query method. We analyze the limitation of the traditional query expansion methods on dealing with semantic, and present 3 semantic patterns of user query and relevant methods to expand the ontology concepts and properties into user queries, which is different from the normal ontology-based expansion. The method use different techniques, such as concept-connected graph and concept hierarchy of ontology, to expand different parts in query, and annotate the issue from semantic view.2. Propose a mothed on ontology-based semantic annotation and clustering on documents, which deals with the representation of documents in semantic retrieval and the division for the centers of the semantic clusters. Based on the semantic preprocessing of the documents, we extract instances to annotate documents; furthermore, utilizing the concept-connected graph, we plot the centers of the semantic clusters. With semantic clustering based on above procedure, we can find the right cluster of user query quickly - it provides effective document management in semantic level.3. Propose and implemente an ontology-based semantic retrieval model. In order to overcome the shortcomings of traditional vector space model in dealing with semantic, we study the problems including statistical method in weight of semantic...