Research On The Application Of Text Mining Technology In Public Opinion Analysis

Public opinion reflects the attitudes,beliefs,and emotions by general public towards certain social events,phenomena and problems,and it is also accompanied by the dissemination and interaction of these opinions.With the development of information technology and the popularization of the Internet,the information dissemination channel has changed from the traditional one-way to multi-direction.People are not only receivers but also producers and transfer stations of information.Net-mediated public opinion plays a stronger character in social public opinion.Given by its advantages in volume,channel,scope and disseminate speed,it has gradually developed into one of the main field for social managers,enterprises,individuals and other organizations to analyze and deal with public opinion issues.Combining with natural language word proeessing technology,text mining discovers valuable knowledge or information from unstructured or semi-structured text information.The analysis of net-mediated public opinion based on text mining has received increasingly concerned in recent years.This paper summarizes the key concepts and technologies of text mining,and emphatically expounds the important methods of text mining,such as LDA model and its visualization,word cloud technology and association rule analysis.Based on the real estate data from July 2018 to September 2018,this paper uses LDA model to extract hot topics and achieve its interactive visualization.Furthermore,the content characteristics of hot topics are depicted by using word cloud technology,and then the hidden association rules are discovered.All of the analysis consequence describes the feature of the net-mediated public opinions of China's real estate in the third quarter of 2018.