[2E4-OS-1a] OS-1 (1)

[2E4-OS-1a-01] Evaluation of Generality on Flaming using Community Bias

Analysis of Long Term Multi-Flaming

〇Fujio Toriumi¹, Takeshi Sakaki² (1. The University of Tokyo, 2. Hotto Link Inc.)

Keywords:Computational Social Science, Social Media, Flaming

Various discourses have been posted on social media, in the case of bursting.
However, it is known that flames are caused by only a part of users.
Therefore, it is highly possible that opinions that spread widely at the time of flames often differ from opinions of the general public.
In this study, we used the known community distribution to determine how the opinion diffused during the flames are differ from the general public by using Kullback–Leibler divergence.
We analyzed the proposed indices for a large-scale flame which continues for a long term.