Real-time detection of students' emotional states in the classroom

摘要

Today, as social media sharing increases, the instant emotional state of the students changes frequently. This situation is greatly affecting the learning process as well as the motivation of the students in the classroom. Sometimes, it is insufficient for the educator to observe the emotional states of students. Therefore, an automatic system is needed that can detect and analyze the emotional states of students in the classroom. In this study, an auxiliary information system, which uses image processing and human-computer interaction, has been developed that would be used in the field of education. In this system, the dataset obtained from the students' faces was tested using various machine learning algorithms. As a result, the accuracy of this system was found as 97.15% using support vector machine. This system is aimed to direct the educator to communicate with the students and to increase their motivation when necessary.