Improving Students' Performance Prediction using LSTM and Neural Network

Educational data mining utilizes information from academic fields to develop renewed techniques and spot unusual patterns to gauge students' academic achievement. Evaluating student learning is a complicated issue. Data mining in this field enables to predict students' performance to recommend performance in universities. Therefore, the current authors have recently seen the rapid growth of data mining and knowledge extraction as tools used by academic institutions to optimize student learning processes. Here, a method based on a certain kind of artificial neural network called Long Short Term Memory recurrent neural network for prediction will operate. The proposed approach tries to use the educational characteristics of different people to predict the best educational process future educational.  It career for students and thereby take steps to improve the effectiveness of the educational system. For comparison, one of the newest algorithms presented in this field was implemented using the proposed technique. The evaluations' findings were performed in the form of two scenarios with different data sizes and different amounts of test and training data. For the evaluation, the dataset taken from an online educational system was used. The evaluation results are presented in the form of four well-known criteria precision, recall, accuracy, and F1, which demonstrate the superiority of the proposed method.