The Semantic Information Automatic Generation

Appearance of the Internet has made great convenience for people. But with the massive growth of information, a lot of useful information has been inundated so that finding the information becomes more and more difficult. To make the Internet service more individuation and intelligent, Tim Berners-Lee, father of the World Wide Web, proposes the concept of Semantic web. In Semantic web, information is represented by Ontology so that the machine can understand the Web information; thereby it is possible to realize more intelligent information service.The proposition of Semantic web has also brought a problem that how to represent the present massive information in a structured form? If all the work has to be done by hand, it will cost plenty of time and energy. Combining with existing information extraction and Semantic Web technology, this paper explore a new technology for Semantic Web and apply it to a travel information service system (TBJ Traveling in Beijing). This technology is automatically transforming the traditional web information to semantic information and store the semantic information in the semantic web required structured form.In this paper, the author analyzes the insufficiency of the present web and the reason why the semantic web appears. According to the characteristic and requirement of TBJ, the article proposed the algorithm and the system architecture of the semantic information automatic generation, and applied it to the TBJ system. The semantic information automatic generation system is mainly composed of three module, online information acquisition, semantic information generation and semantic information representation.In the content related extraction, the author proposes a method using the semantic similarity. Two extraction approaches are adopted in the process to make good use of the semi-structured characteristics, one is web document structure related, and the other is content related, which brings good precision.