Design And Implementation Of The Public Opinion Monitoring System Based On Semantic Web

With the rapid development of network information technology, especially the appearance of social networking services, there is an advancement of free speech. But using the network to spread rumors or to distort the truth can cause a harmful effect on society. Consequently, in order to protect the network environment from these abuses, it is essential to use information technology to accomplish the automatic monitoring of public opinion on the network. Most of the public opinion monitoring systems running on the network use a process which compares one or more key words used by the public on the network to determine the relevant public opinion information that is abusive, but those systems have a lower accuracy rate.According to actual needs of Lanzhou Petrochemical Industries Company network public opinion monitoring, this thesis designs and implements a public opinion monitoring system based on Semantic Web. This system employs the technology of Semantic Web to public opinion, collects domain knowledge of public opinion, and builds public opinion ontology and emotion ontology. And the character of the ontology inference can be used to analyze the information of public opinion ontology. Through setting the strength and the level of the emotion ontology and judging the subjective tendency of the information, we can recognize the efficient, accurate impact of public opinion.The two main functions of the public opinion monitoring system are information collecting and analysis of public opinion. The system design has two layers of structure: the information collecting layer and the public opinion processing layer. The information collecting layer has a “Web Spider” module, a “Web page purification” module and a “Chinese Word Segmentation” module. Their functions are widely collecting network information, extracting main contents from the collected web pages, and word processing to the pure text information extracted. Public opinion processing refers to information collected by the ontology library of the public opinion to accomplish the reasoning work of the public opinion and to create the final report of public opinion.This thesis mainly completes the detailed design and implementation of the public opinion monitoring system based on Semantic Web according to the requirements analysis. Now this system has been put into operation. It can achieve the required accuracy rate and contribute to the stability of work at Lanzhou Petrochemical Industries Company.