Presentation information

General Session

General Session » J-5 Web intelligence

[1L4-GS-5] Web intelligence: Social data analysis (1)

Tue. Jun 9, 2020 3:20 PM - 5:00 PM Room L (jsai2020online-12)


4:00 PM - 4:20 PM

[1L4-GS-5-03] Analysis on incitement degree of media company social media account based on social porn hypothesis

〇Takeshi SAKAKI Sakaki1,2, Fujio Toriumi2 (1. Hottolink, Inc., 2. The University of Tokyo)

Keywords:Social Media Mining, Computational Social Science

In recent years, due to the spread of web and SNS, various problems related to information diffusion such as the spread of fake news and the occurrence of echo chambers have become apparent. Under these circumstances, techniques to evaluate the reliability and bias of information and information sources have been being developed. In this study, we focus on the incitement as one of information source features (hereinafter, media), and propose a quantitative index to evaluate it. In this study, we assume that there is "social porn(information that people want to spread and share immediately), and try to propose a quantitative index to evaluate the quality of information source (hereinafter, media), based on the hypothesis. We define {t user response time for posts, which is How fast the user spreads the information which he/she received, as a clue to estimate the degree of incitement. We proposed an index that summed up user reaction time for each media as an index indicating the incitement. As a result, it was suggested that the proposed index tended to have a unique value depending on the media, irrespective of the article and the degree of article diffusion. In the future, we will conduct large-scale user evaluation experiments to further verify the validity of the proposed index.

Authentication for paper PDF access

A password is required to view paper PDFs. If you are a registered participant, please log on the site from Participant Log In.
You could view the PDF with entering the PDF viewing password bellow.