Research And Application Of RDF Inference Mechanism

A big flaw of Internet is that it doesn't provide enough semantic information abount web resources. Computers cannot understand what the information mean. This is why the architecture of WWW is transforming to semantic web. In 2000, W3C proposed the semantic web as the next generation of WWW. Resource Description Framework (RDF) plays an important role in semantic web. The goal of RDF is to implementing sematic description of web resources.We have done a lot of research on the semantic web and RDF inference mechanism and proposed a information retrieval framework based on RDF inference mechanism. This framework will solve the problem of semantic description of web resources and make computer easy to identify synonyms or relative retrieval. This framework is composed of semantic modeling of certain domain, semantic description of resources, extension search of semantic model, and etc.This paper emphasize on RDF inference mechanism, which is the fundamental of sematic retrieval. RDF inference is carried out on a triple based model theory. First we build the semantic model of certain domain by ontology, and then add the semantic description about resources by RDF. The incorporation of ontology and RDF forms the knowledge base. That can support some limited inference according to RDF entailment rules. But there are some limiations of RDF semantics. For example, RDF schema can only organize the vocabularies in typed hierarchy and can't specify some characteristic of properties, such as transitivity, symmetry. In order to remedy the limitation, we dynamic extend the semantic model through domain extension rules, which are predefined by ourselves.We have developed a semantic retrieval system on computer network field and expected to provide a intelligent learning resource retrival service. This system tries to solve problems that ordinary systems cannot do, such as identify synonyms, relative retrieval. The result of experimentation proves that this retrieval framework makes semantic of resources sufficiently emerged and the good feasibility of search strategy.