| With the popularity of the Internet and the rapid increase in the number of Internet users,research on the impact of online news has gradually become a hot spot of concern.Online media is more complex than other mass media.Its timeliness,authority,integrity and interactivity make the audience more extensive,but the problem comes with it.The influence of the news media on publishing news propaganda is still unclear.The amount of news reading published by the online news media was not satisfactory,which ultimately led to the failure to change the direction of public opinion in a timely manner,thus affecting social peace.In this thesis,a model that reflects the effects of online news media propaganda is proposed.This model extracts relevant information from news web pages,and then use text processing and other related technologies to analysis of the impact of online news media.Explore the relationship between it and the topics of concern to netizens on the popular discussion platform,and make feasible suggestions for relevant departments.Firstly studies the influencing factors affecting the effect of publicity on the news posted on the media website,and extracts the number of responses to the news pages that can be quantitatively analyzed,the number of reprints,the published source website,and the release time of news.Use reptile technology to build a initial model.Secondly,quantify the influencing factors of these capture,and the statistical analysis of the response rate is used to obtain the response rate ratio table as the reference.The source impact factor is calculated by using the daily average number of visitors and the average daily page of the alexa website.The time-influencing factors are obtained by using the definition of news recession time proposed by the relevant papers.The reprint rate is indicated by The source website is obtained through the VSM model,the MD5 algorithm,the TF-IDF feature weighting technique,the word co-occurrence method,and the HITS algorithm.Finally,the initial model is optimized by correlation analysis and regression analysis.The results of optimization and Baidu ranking data are used to analyze the impact of the publicity of online news media content and the differences between the publicity effect and the topics concerned by netizens on the public discussion platform.This thesis compares the results of the news and publicity effect with the hot spots of Baidu's rankings.It is found that the factual news of the website news media is mostly in the content,and more biased into the news of politics,science and education,economy,politics and law.Today's hotspots in Baidu's rankings are biased towards commentary events,reports of social events,problems,and styles involving the daily lives of the people,which are likely to cause discussion,judgment,and controversy in society.From the order of the two,the rankings of news and publicity influences can basically be consistent with Baidu's ranking of today's hotspots.They are also in the direction of social and critical events,but the order of factual news is not satisfactory.Factual news,the greater the difference in ranking,and the more commentary topics,the smaller the difference in the order of ranking. |
PDF Full Text Request