A machine learning approach to Predict the Engineering Students at risk of dropout and factors behind: Bangladesh Perspective

Dropout rate in Bangladeshi universities getting high day by day. Especially in engineering subjects. Massive number of students taking various engineering subjects for their under graduation. However, the completion rate is low. There can be mainly two types of reasons behind dropout- Academic or personal reasons. The target of this study is to find the factors behind the high dropout rate in Bangladeshi universities engineering subjects and also to detect risky profiles for dropout so that their dropout can be prevented. Current and previous student's data is analyzed to find the factors and students at the risk of dropout, which can be useful for developing new strategies in the education system by universities or other educational authorities. SVM, random forest, neural network, etc. were used for creating the prediction model.