A Study on Performance of Mathematics, Programming, and Practical Courses Among Female Students from Technical Education Using a Deep-Learning-Based Interpretability Framework

The Mathematics, Programming, and Practical courses are herculean task for most of the students pursuing their study in the field of engineering and technology. This study has explored the performance of female students in these three categories pursuing Bachelor in Technology (B.Tech) in Computer Science and Engineering (CSE). A deep learning (DL)-based prediction model is proposed to evaluate the academic performance of students at the end of their B.Tech days measured by cumulative grade point average (CGPA). The model is based on two deep learning methods—convolutional neural network (CNN) and bidirectional long short-term memory (Bi-LSTM). The overall predicted results are interpreted using an interpretable supervised clustering framework. A comparative study is performed using two baseline models—long short-term memory (LSTM) and gated recurrent unit (GRU) to check the acceptability of the proposed model. The empirical results confirm that the losses encountered in the proposed model in terms of mean square error (MSE) and mean absolute error (MAE) are less than the other two base-line models that improve the overall performance. The results mirrored from the proposed model has established the fact that the trend of female student’s performance in Mathematics, Programming and Practical courses is comparatively lower compared to the male students.