A Real time Application for Student Stress Level Prediction in education sector using Bayesian-KNN (NB with KNN) Algorithm

College understudies are enduring from numerous mental wellbeing issues counting mental push, somatization, fanatical, interpersonal affectability, discouragement, uneasiness, threatening vibe, fear, distrustfulness and mental disturbance, which can bring a parcel of negative impacts to them. Clearly, the mental prosperity issues of college understudies not because it were particularly impact their have advancement, but as well impact the consistent quality of the campus. We require a system to handle the understudy mental wellbeing issues. Here we are mainly targeting on predicting levels of stress . We can observe several variables related to mental issues such as age, sex, work culture, assignments, conflicts within family, companions issues, participation, education etc. Utilizing machine learning strategies we anticipate understudy stress and levels. In the project we used two administered learning algorithms such as knn and bayes algorithm and applied combination of these for prediction and got better results with an accuracy 94.5% which is more compared to 80.5% and 85.7% for KNN and bayes respectively, we named that combination as hybrid algorithm. Moreover our system is going to provide some useful suggestions based on the stress levels of understudy. In this proposed system we develop automation for education sector for student stress and levels prediction.