Font Size: a A A

Analysis Of The News Text Frame Of "China News" Based On Text Mining Method

Posted on:2021-04-11Degree:MasterType:Thesis
Country:ChinaCandidate:Y J WangFull Text:PDF
GTID:2438330623972531Subject:Communication
Abstract/Summary:PDF Full Text Request
In the study of communication,the study of news framework has been enduring.This paper attempts to use the text mining method to construct and judge the framework of China news.First of all,the research demonstrates the theoretical rationality of the framework analysis of text mining,and compares content analysis with text mining to find common features.Furthermore,the characteristics of news text and human language understanding support the framework analysis of text mining.After the demonstration,the paper begins to construct and judge the framework of the Chinese news.In the framework construction,the technology of keyword frequency statistics,named entity,keyword co-occurrence network is used.Based on the interpretation of "new text" of the co-occurrence network of person name,place name and organization name,a news framework based on subject words can be constructed.According to the results of the analysis,the framework of specific subject words is constructed.After that,the theme words are classified into international news,domestic news,Hong Kong,Macao and Taiwan news,the paper summarizes the framework of China News in three types of news.Finally,through interviews,it proves that the function of China News framework is invalid and its reasons.In the framework determination,based on Semetko's scale and its five general frameworks,this paper analyzes the dimensions of the scale,and constructs a vocabulary of five general frameworks.The significance of the framework is discussed by adding up the word frequency of different framework vocabularies.From this,we can know the tendency of China news,that is,it tends to be macro,economic and introductory news,rather than specific,close to the people and in-depth news.
Keywords/Search Tags:text mining, content analysis, news context frame
PDF Full Text Request
Related items