A comprehensive methodology for detecting stress and performance of students using machine learning

Abstract

Nowadays, there is a big problem with mental stress, especially in young people. The time period that was formerly thought to be the most carefree is currently under a lot of stress. Today’s increased stress causes a variety of issues, chief among them depression and suicide. In this experiment, we are measuring student emotional stress and assessing their academic achievement.The often-unnoticed impact of exam pressure, employment pressure, or recruitment stress on the student. The work conduct a study on how these elements impact a student’s psyche and show the prevalence of stress among students.An automated system ”Student Eval” which collects input from user and the model build using machine learning algorithm such as SVM on the dataset and model is saved.The front-end application is to gather new user features and pass them on to the stress prediction model. In addition to academic performance, clubs participation, competition participation, and achievements, teachers can also keep an eye on students accomplishments.The Teacher monitors their kids stress levels and give the right guidance and also examine the causes of stress, such as the amount of work given to the student, how long it takes them to finish it, and unfinished tasks.To identify stress and emotion in real time, a pre-trained model using CNN is used. Using opencv, a video of students is collected frame by frame. Movements of the lips and brows are used to calculate stress levels..The proposed system performs better than the existing system since most of them rely on wearable sensor data, but the proposed technique is less expensive because it uses data from a web application. A small institution or school system can use this strategy to examine the stress and academic achievement of its students.