Analysis of students's behaviour is important for instructors because it is one of feedbacks from students that let the instructors to understand their students. This can enable the instructors to be able to improve their teaching methods or materials. Examples of Tracking students' behaviour in the class are emotion, attendance, and action. Currently, it is very difficult and laborious to observe these behaviours when there are a large number of students in a class. Therefore, we propose a system called "Classnalytic" to assist instructors to track students' behaviour in their class. The proposed system utilises computer vision and machine learning techniques to tackle the problem.
The proposed system can identify students and perform attendance tracking using a camera installed at the front of the classroom