Characteristics of Micro-Expressions Occurring During Confusion

• Published in: Proceedings (International Conference on Machine Learning and Cybernetics.) pp. 327 - 332
• Main Authors: Hata, Kotmi, Takano, Hironobu
• Format: Conference Proceeding
• Language: English
• Published: IEEE 20.09.2024
• Subjects: Cybernetics; Emotion Estimation; Estimation; Lips; Machine learning; Micro-expression; Support vector machines; SVM
• ISSN: 2160-1348
• DOI: 10.1109/ICMLC63072.2024.10935167

Asking questions about unclear points in university lectures and exercises is desirable because it enhances the lecture out-comes; however, Japanese university students are noted for their low propensity to ask questions, often due to reasons such as em-barrassment. Leaving doubts unresolved could hinder the learners' growth. Therefore, it is speculated that if instructors can estimate when learners have doubts, they can potentially extend support, leading to problem resolution. In this study, the state of having doubts is defined as a state of confusion, and the aim is to estimate the state of confusion using micro-expressions, which are facial expressions that manifest even when attempts are made to suppress or conceal them. We constructed a dataset of micro-expressions observed during confusion. Furthermore, using existing micro-expression datasets, we evaluated this dataset based on its ability to discriminate confusion from other emotions.