Studies On Online Public Opinion Monitoring And Related Technologies

With the rapid development and popularity of the internet,online public opinion monitoring has become more and more importantthan ever before. Accidental incidents often cause hot topics on theinternet and are widely spread. The internet has become a major channelthrough which the government get to know the public opinion. How toget the public opinion and how to process it have become a challenge tothe governments at various levels.In this thesis, we discuss online public opinion monitoring andrelevant technologies. For this purpose, the author has designed aninformation retrieval system and a text clustering system as experimentplatform for the study of to techniques in Chinese information processing:Chinese word segmentation and text clustering. The major resultsobtained includes the following.1．For faster processing of Chinese word segmentation, we haveproposed a word segmentation method based on balanced binary treestructure of the dictionary and another structure using double-characterhash table and frequency information. Experiments show that ourmethods are effective.2. we have proposed an improvement on the K-means clustering method and showed its effectiveness by comparison experiments.