情報処理学会 第80回全国大会講演要旨

Heart Rate Variability (HRV) has emerged enormous potential on cognitive performance evaluation as taking insight into the autonomic nervous system. In this study, we propose that HRV of presenters can be used to effectively evaluate one of the cognitive performance: discussion performance, which consists of several Q&A statements. To validate our opinion, noninvasive device Apple watch was used to collect presenters' HRV, three machine learning models: Logistic Regression, Support Vector Machine, Random Forest has been generated and discussed. Comparative experiments were performed to evaluate discussion performance using traditional semantic data of Q&A statements alone and the combination of HRV and semantic data. We also confirmed the robustness of HRV features on the new dataset.