A Novel Ensemble Stacked Learning Frameworks for Effective Evaluation of Ideological and Political Courses

With an explosive intrusion of Internet, several universities and educational institutions have adopted the online evaluation system of teaching to collect student feedback thereby improving the students' calibre. In the recent advancements of machine and deep learning algorithms, invaluable insights from these evaluations can be extracted, offering the intelligent and robust solution for improving the teaching and teachers' quality. In this regard, this research paper proposes the novel ensemble of Bi-directional Long short term memory (LSTM) and Bi-Layered Machine Learning algorithm covering both subjective and objective evaluations suitable ideological and political courses. The suggested model first pre-process the data and subsequently adopt the ensemble model to evaluate the contents which tends to predict the sentiment feedbacks. The comprehensive experimentation has been conducted using the OULAD datasets and performance measures like accuracy, precision, recall and F1-score are calculated. To prove excellence of model, validation parameters are compared to other existing baseline learning models. Evaluation results demonstrates the ability of the model to achieve the accuracy of 98.4%, precision of 98.2%, recall of 98.1% and F1-score of 98% respectively. This intelligent system effectively cultivates the evaluation data, providing cognitive tool for enhancing the teaching quality.