Blockchain Framework for Learner Performance Prediction using Life-Brain Storm-based Light GBM Coupled Neural Network

Abstract

E-learning is one of the dominant applications of digital techniques in the educational platform. Tutors can effectivelytailor their instruction to each student by using the automatic identification of the student's learning styles. Nowadays Deep learning techniques provide the preferable predictive model in the e-learning platform. Hence, this research article provides the prediction of the learner’s performance by using the Life-Brain Storm (Life-BS) based LightGBM coupled Neural Network (NN). A significant part of the research lies in the tuning of the hyper-parameters using the proposed Brain rule selection algorithm, which boosts the accuracy of the classifier. Furthermore, by lowering the dimensionality of the data, the feature extraction approach is developed in this study to reduce the computational complexity of the prediction framework. The suggested Life-BS-based LightGBM coupled NN model is shown to be effective by the experimental assessment, which yielded the lowest RMSE as well as the MSE for courses 1, 2, and 3, respectively. In addition, the evaluation metrics such as MAE and Kappa scores achieve better results for course-1, course-2, and course-3 respectively. Use of blockchain, including kappa score also in performance metrics along with Life-Brain Storm based LightGBM coupled Neural Network proposed learner performance prediction model are the keypoints of the presented work.