Rating the Credibility of Online News Stories

Abstract

Abstract The internet has become one major platform where people get information regu- larly. News reading on the web has increased throughout these years. At the same time number of visits for a news website has increased. Hence today people try to manipulate information on the web in many ways. So here comes a new problem called fake news which can do considerable in uence on events such as elections, natural disasters etc. With the involvement of the social media, this problem has become even bigger, because the information in social media is not monitored can be manipulated easily. And social media has the power of spreading information in less amount of time. Therefore identifying credible information from the web has become really important today. In this thesis, we propose a novel approach for ranking the credibility of the web- site/source based on their behavior on the web. In order to determine the credi- bility of online news stories, rst we need to determine the credibility of the news website. Hence this research is focusing on determining the credibility of the news website. So the approach followed in this research is based on the behavior of news stories on Twitter. The literature discusses three main user in uencing factors in Twitter. They are In-degree, Mentions and URL Recommendation. So based on these factors three di erent models are developed that produce credibility rank- ings for news websites. And nally, a survey is conducted with experienced and reputed journalists in Sri Lanka to evaluate credibility ranking values produced by the models. According to the experiments carried out, it indicates, the fac- tor URL Recommendation is the most in uencing factor of news credibility in Twitter. So this research contributes a Credibility Network Model that produces credibility ranking values for Sri Lankan news website by considering the factor URL Recommendations in Twitter.