A Study On Ontology-based Knowledge Retrieval

The increasing number of information resources and the developing level of people's information demand in modern society give information retrieval technology a great challenge. New ideas and methods are introduced into the field to reform its theory and technology. So comes one ofthe hot topics--knowledge retrieval (KR), which has close association with artificial intelligence(AI). Ontology-based knowledge retrieval, which is discussed in this paper, just embodies this association.Ontology is an advanced technology of knowledge representation in AI. It uses concepts and relations between concepts to describe abstract facts and build models. In KR system, ontology can be used to describe deep knowledge content in a structural way. Knowledge actions, including searching, filtering, extracting and integrating, are executed intelligently based on the models. This paper firstly studies the basic theory of ontology, emphasizing on ontology-based knowledge representation, organization and retrieval methods. Then it discusses the generic structure and running mechanism of ontology-based KR system. Finally it explores an example to show how to use ontology in a practical system. The paper contains five parts:Part 1 introduces the fundamentals of ontology, including its basic knowledge such as definitions kinds purposes etc, and its practice knowledge such as building principles and methods, ontology languages and tools etc.Part 2 studies the ontology-based knowledge representation technology and analyzes how to give an abstract description of knowledge content using representation elements. The main content of this part includes how to acquire concepts, how to define knowledge in ontology and how to integrate different ontologies.Part 3 studies the ontology-based knowledge repository. It puts emphases on the organization of four typical kinds of knowledge base in KR system, including language knowledge base, expert knowledge base, domain knowledge base and fact knowledge base.Part 4 studies ontology-based KR system model. It first proposes a generic structure of the system, and then analyzes the function and principle of every module in the structure. Furthermore, it discusses the mechanism of ontology-based KR reasoning and personalized information service.Part 5 anatomizes KRetrieval, a KR system model developed by us in the project. By specifying the model's structure, function and features, it gives a picture of the use of ontology in practice.This paper is one of the product of "information visualization and knowledge retrieval"--the important science project of ministry of education (Project ID: 02JAZJD870004) .