[2F1-OS-20] OS-20

9:30 AM - 9:45 AM JST(12:30 AM - 12:45 AM UTC)

[2F1-OS-20-03] Committee analysis using the chronological change of affiliate-based turn-taking for human capital visualization

〇Ekai Hashimoto^1,2, Kohei Nagira², Takeshi Mizumoto², Shun Shiramatsu¹ (1. Nagoya Institute of Technology, 2. Hylable Inc.)

Keywords:

Quantitative conversation analysis,Communication,Organizational analysis,Meeting minutes

When analyzing the organizational structure of regularly meeting committees, content-based analysis of utterances is commonly employed.
However, for confidential internal meetings, conducting such an analysis is difficult because neither the content nor the outcomes can be disclosed externally.
To address this issue, we have proposed methods that do not rely on linguistic information: one based on speech amount and another based on turn-taking information.
In the turn-taking–based approach, we focused on turn-taking between attributes of the dominant speaker and classified committee characteristics from the resulting network structure.
In the speech-amount–based analysis, we showed that organizational changes can be captured by examining temporal variations.
Therefore, in this study, we extend the turn-taking–based approach to per-meeting analysis and investigate time-series changes in a committee.
As a result, we identified multiple meeting states within a single committee.