Description Logics Based Knowledge Representation And Reasoning For The Semantic Web

World Wide Web has become one of the important channels for people's access to the information and services. But as a collection of paper media, most of its contents are well designed for human, the computer can not understand or handle them automaticly so far. Therefore, although the World Wide Web is a virtually unlimited potential in the information resources, its potential has not been fully applied. To solve this problem, researchers put forward the concept of semantic Web.Semantic Web is first proposed by Tim Berner-Lee who is the founder of the Web, which is described as"it is the expansion of the current Web with the definition of good semantics, it can make the computer and man work together better."The goal of semantic Web is to make the computer understand the information on the Web. In order to achieve this goal, researchers of the semantic Web have developed many new technologies, and put forward a series of technical standards. Tim Berners-Lee has been committed to the semantic Web technology research and raised a level of the semantic Web model. This model has been widely recognized by the researchers of the semantic Web. This model is composed by seven lays: Unicod and URI layer,XML+NS+XML Schema layer, RDF/RDF Schema layer, Ontology layer, and Logic, Proof and Trust layer.Knowledge representation is the basis of Semantic Web, knowledge representation means the knowledge should be coded into an appropriate data structure. In order to facilitate knowledge management analysis and reasoning, knowledge representation methods are typically required simple, clear and easy-to-understand, at the same time able to expand knowledge flexibly. Traditional knowledge representation methods include several ways that are logic, expert system, framework and semantic network etc. The above-mentioned are usually concentrated. In a word, to each of common concept , people must use the same definition. The knowledge representation of semantic Web is to build a date and knowledge of many-to-many exchange platform, its core content is to establish a clear semantic information space.The greatest challenge of semantic Web is how to provid a language, which is able to describe the data, as well as the data reasoning rules based on it, and allow any knowledge identification system output to the Web, then the logic of the Web is strengthened.So far, the important technologys that solve this mandate including XML, RDF, and ontology, etc. XML is the cornerstone of the semantic Web which provides a way that is flexible, general and structural. RDF is a definition of a simple data model which can represent the relationship between the expression data. According to defining attribute, ontology establishs a classification hierarchy,different concepts can be distinguished and organized. At the same time, the definition can be contacted mutully, so the semantic space of concepts is established. Ontology is the foundation of semantic Web which resolves information sharing and exchanging, which plays a very important part in the semantic Web.Description Logic is a formal tool based on the objects knowledge representing. It is a logic formula that used to describe and reason the defined knowledge. It is a decidable subset of the first order predicate logic, and it can offer decidable reasoning function. At last it is the logic foundation of semantic Web's Ontology layer and plays a very important role in Semantic Web's development.In this paper, traditional methods and the semantic Web knowledge representation methods are firstly introduced. Secondly, shortcomings and limitations of these methods are analysed. At last, the semantic Web knowledge representation methods based on describtion logic are studyed and investigated.There are two main contributions in this paper. Firstly, the relationship between F-Logic(Frame Logic) and an expressive and decidable description logic is given, which basised on Mira Balaban's work. The expressive and decidable relationship demanstrates that a description logic SHOIN(D) ontology can be transformed into an F-Logic ontology equivalently. Which give a way of how to integrate description logic and rules and a inference algorithm is given. The correctness and the complexity of the algorithm are proved. Secondly, based on Sown and Baader's work, a way that how to transform the knowledge which is represented by a descidable concept graphs into a description logic ontology is given, as well as the validity of such a transformation. So the existing automated reasoning algorithm of description logic can be used to reasoning of concept graphs', and the validity of such a transformation is also given. This work fetchs up the inadequate part of Sown and Baader's study and provides a solid theoretical foundation for the application of concept graph in the semantic Web.