Quality Evaluation of Ideological and Political Education Using Restricted Inverse Learning Strategy Based Gazelle Optimization Algorithm with Back Propagation Neural Network

In Ideological and Political Education (IPE) system the teaching quality is crucial to analyze the comprehensive level of university to enhance the teaching process. However, the existing machine learning and Deep Learning (DL) based evaluation models used for estimating the teaching quality of IPE system failed due to adopting the environmental factors and prediction error lead to inaccurate results. To overcome this problem, an optimized DL model namely Restricted Inverse Learning strategy based Gazelle Optimization Algorithm with Back Propagation Neural Network (RIL-GOA-BPNN) is proposed to estimate teaching quality of IPE system precisely. The natural behaviors of gazelles in GOA like stotting, evasive running lead to adopt the frequent changes of education factors that has vital part of teaching quality of IPE system. Moreover, the proposed BPNN model that learns complex, nonlinear relationships between various evaluation factors to enhance the accuracy and reliability of teaching quality. The GOA has certain limitations which affect accurate evaluation results, thus a RIL strategy is utilized to solve the problems to enhance the teaching quality evaluation by adopting different education patterns effectively. Experimental results of proposed RIL-GOA-BPNN teaching quality evaluation model achieved 0.654 and 8.159 of Mean Squared Error (MSE) and Mean Absolute Error (MAE) for 100 epochs which is less than state-of-the-art-methods.