Estimation of Individual Contribution Ranking by Group Assessment on the Subject of Manzai Performances

Abstract

In recent years, there has been a growing need for individual contribution evaluations. However, there is a problem in that group assessment do not provide information on individual contributions. In this study, we focus on information on Twitter and aim to estimate the contribution of each performer by using the number of likes on the promotional posts of a manzai theater as a feature value. Specifically, we propose a novel method that applies the TrueSkill\texttrademark algorithm. In an experiment to evaluate the accuracy of the proposed method using the number of likes on the promotional posts of a comedy theater, it was shown that the proposed method improves the accuracy when the number of likes by manzai fans is excluded. In addition, it was confirmed that the values calculated by the proposed method correlate with the results of a manzai competition, suggesting that the proposed method is effective.