Font Size: a A A

The Key Technology Research And Implementation Of Network Public Opinion Monitoring System

Posted on:2017-04-12Degree:MasterType:Thesis
Country:ChinaCandidate:H M TuFull Text:PDF
GTID:2308330503979167Subject:Computer Science and Technology
Abstract/Summary:PDF Full Text Request
Internet based public opinion analysis has been widely used, but with the emergence of new communication platform, the Internet public opinion analysis is also constantly expanding and improving. Today, Internet public opinion monitoring has been covered by micro-blog, We Chat, forums, blogs, post bar, news and other aspects. Micro-blog as one of the most popular social networking tools in recent years, its spread fast, interactive, and other characteristics, has been developed as an important platform for information exchange and sharing of the internet. At present, there are still some deficiencies in micro-blog public opinion monitoring. This paper focuses on the analysis and research of micro-blog public opinion monitoring related technology from micro-blog. The main research work and research results of this paper are as follows:According to the characteristics of micro-blog’s short text, the micro-blog short text information pretreatment program is analyzed, and the "unkno wn words" in micro-blog are identified. According to the special form of micro-blog text, extracted micro-blog topic text, after filtering and calculating the value of mutual information and other steps to identify micro-blog new words. The establishment of OOV lexicon, thus improving micro-blog text segmentation effect.The related theories and technologies involved in the monitoring of the traditional public opinion are studied, and the difference between the traditional text and micro-blog short text is compared. In-depth analysis of the limitations of traditional clustering algorithms in microblogging short text clustering, using VSM vector space model and LDA document theme generation model to optimize the performance of K-means clustering algorithm, optimized K-means initial center establishment method and text similarity calculation formula and to improve the accuracy of clustering.Internet media has long been deep into the daily life of the people, the highly active users of domestic and foreign any hot events can immediately produce public opinion. It has brought great convenience to people’s communication, and it has also become an important carrier of social public opinion. Therefore, public opinion monitoring is very necessary for the state government and the network supervision department.
Keywords/Search Tags:Internet, Public Opinion Monitoring, Weibo, Unknown Words, Text Data Clustering
PDF Full Text Request
Related items