[1L5-GS-5] Web intelligence: Social data analysis (2)

[1L5-GS-5-01] Analysis of Behavioral Changes in News Service Focusing on Popularity

〇Atom Sonoda¹, Hiroto Nakajima², Fujio Toriumi¹ (1. The University of Tokyo, 2. Nikkei inc.)

Keywords:online news, recommender system, diversity changes

Information is transmitted through websites, and the immediate reaction to various information is required. Hence, the efforts for readers to select information themselves have increased, which leads to the further improvement of recommendation services that can reduce such burdens. On the other hand, it has been pointed out that there are problems such as filter bubbles and echo chambers that provide only biased information to users due to excessive recommendation. We aim to quantitatively evaluate these user behaviors from log data. So far, we have discussed user behavior change based on the diversity of article categories. In this paper, it was shown that the change of user behavior can be grasped in more detail by using information about article popularity and reporters in addition to article category information.