chapter  Chapter 12
14 Pages

Detecting Singular Data for Better Analysis of Emotional Tweets

WithKiichi Tago, Kenichi Ito, Qun Jin

In this chapter, the authors introduce analyses that consider outlier users on Twitter. They examine the influence of emotional tweets on user relationships using some emotion dictionaries. There are many studies that examine users’ emotional expressions and relationships on Twitter. Social networking services (SNS) such as Twitter have been used as a platform by people across the world for easily posting messages. S. M. Horng analyzed the relationship between the user behavior in SNS and the indicator of Google Analytics. In order to have a better analysis and obtain a more accurate result, it is necessary to detect singular data and exclude them. The quality of a dataset may be further improved by excluding singular data carefully. The authors analyze the top seven users who had the most emotional tweets when evaluating emotion using the emotional word dictionaries.