Automatic recognition of student emotions from facial expressions during a lecture. (April 2020)

• Record Type: Journal Article
• Authors: Tonguç, Güray; Ozaydın Ozkara, Betul
• Is Part Of: Computers & education. Volume 148(2020)
• Publication Date: 2020-04
• Subjects: Data science applications in education -- Applications in subject areas -- Architectures for educational -- Technology system -- Improving classroom teaching -- Teaching/learning strategies; Education -- Data processing -- Periodicals; Education -- Periodicals; Computers -- Periodicals; Computer-Assisted Instruction -- Periodicals; Éducation -- Informatique -- Périodiques; Electronic journals; 370.285
• DOI: 10.1016/j.compedu.2019.103797
• Languages: English
• ISSNs: 0360-1315

Abstract

Abstract: Emotions are of great significance in education as in all areas of human life. It is acknowledged that, although the cultures of people, the environment where they live, and the language they use vary, emotions that are evaluated as universal do exist. In this study, we examined the changes in the emotions of 67 students during the lectures of Basic Information Technologies who studied in three different departments in a state university in Mediterranean region. The facial expressions of the students were analyzed and digitalized in terms of the feelings of disgust, sadness, happiness, fear, contempt, anger and surprise by means of a software developed through Microsoft Emotion Recognition API and C # programming language. During the lecture, we studied how student emotions varied and whether this change was statistically significant according to their departments, gender, lecture hours, the location of the computer in the classroom, lecture type and session information. The lesson plan consisted of three stages: introduction, activities and closure. The significance of the difference in emotions and the relationship between emotion change and achievement were examined according to three different stages. The research showed that feelings of contempt, anger, fear and confusion increased, while feelings of happiness, sadness and disgust decreased in the first stage of the lecture. It was determined that the feeling of happiness increased rapidly, whereas all other emotions decreased during the closure section of the lecture. We found that only the feeling of sadness displayed a significant difference according to the departments. It can be suggested that the equipment that provides instant feedback to the lecturer by monitoring the students' emotions automatically throughout the lecture can be made available throughout the educational institutions, thus contributing to the increase in the quality of education. Highlights: Students' emotion changes during the course can be analyzed with computer aided facial expression analysis systems. During the first 4, 36 and 46 min, there are sharp changes in mood. Emotion changes are based on gender, department, sitting order, session, etc. does not depend on factors. Emotional changes are similar to the periods in the literature.