Research On Semantic Search And Related Technology

In current century, how to achieve useful information for the users from huge mount of information is one of the main problems confronted with people, and semantic search is a hopeful way to solve it. The rising of the semantic web and the maturing of related theory and technologies, together drive the rapidly-developing semantic search research area. Several key problems in semantic search domain are addressed in this dissertation, and our work in this dissertation can be divided into the following parts.1. Driven by the analyzing the synonymic characteristic of the key words and the intensity of semantic association among the ontology instances, we provide a user's intention analysis scheme, which synthesizes the synonymic relation and the semantic association. The method can improve the accuracy and completeness of the mapping from key word to ontology entities, infer the user's intent.2. Determining the semantic similarity is an important issue in the development of semantic search technology. A scheme was presented to calculate the semantic similarity, which took multi-inheritance of entities, hierarchical structure of the properties and property values into consideration, and then optimized the computing process based on the tree structure of inheritance relationship. The experimental results show that the scheme can calculate semantic similarity more precisely.3. User profiles, descriptions of user interests, can be used by search engines to provide personalized search results. A user profiles analyzing method based on semantic association is presented. This method gains the users' interests by the implicit methods and explicit methods, and creates user profiles by classifying users' interests into instances in an ontology knowledge base, and then propagates user preferences to find users' latent interests by analyzing the semantic association among the ontology instances. It integrates users' current and history preferences to process the search results. The experimental results show that users' latent preferences can be learned accurately and personalized search based on user preference yields significant improvements over the original results.4. As the competition of Web Search market increases, there is a high demand for accurately judging the relations between the web pages and the user's requirement. In this paper, we propose an information retrieval method that tightly integrates description logic reasoning and traditional information retrieval technique. The method expresses the user's search intention by description logic to infer the user's search object, and selects high-quality keywords according to the semantic context of the search object. Further, fuzzy describing logic is introduced to confirm the relations between the web pages and the user's search requirement, and the method to calculate the membership degree of web pages w.r.t the search requirement is presented. A prototype is implemented and evaluated, and the results show large improvements over existing methods.5. The semantic search methods, proposed in the dissertation, had been implemented in "Integrated Information Service System". In the dissertation, we introduced the system, and discussed the way of information provision which integrated call center, LBS (location based services) and semantic search by concrete demo.