Application of Natural Language Processing Technology in Student Evaluation and Feedback in Teaching Quality Assurance System

Abstract

As the on-going reform in higher education teaching deepens and expands, there has been a significant focus on researching educational quality. Enhancing teaching quality is crucial for improving education, with teacher evaluation serving as a vital instrument in achieving this goal. In this manuscript proposes an Application of Natural Language Processing Technology in Student Evaluation and Feedback in Teaching Quality Assurance System (NLTSEFTQ-MORA-RNN) . Initially, the data is collected from the 2 separate universities data set. Then, the collected data is fed to Pre-processing segment. In pre-processing stage, Multimodal Hierarchical Graph Collaborative Filtering (MHGCF) is used to clean the data. Then pre-processed output is given to Mixed-Order Relation-Aware Recurrent Neural Networks (MORARNN)that classifies the teaching qualification such as Leadership Evaluation, Expert Evaluation, Peer Evaluation and Student Evaluation. The weight parameters of MORARNN are optimized using Fennec Fox Optimization Algorithm (FOA). The proposed NLTSEFTQ-MORA-RNN method is implemented and the performance metrics such as Accuracy, precision, sensitivity, specificity, F1-score, and computational time are evaluated. By then, the performance of the proposed technique is executed in the Python platform. The performance of the proposed NLTSEFTQ-MORA-RNN approach attains28.5%, 27.5% and 28% higher accuracy, 25.06%, 25.33% and 22.98% higher Precision and 27.12%, 21.33% and 24.98% higher sensitivity compared with existing methods such as an improved genetic algorithm and neural network-based evaluation model of classroom teaching quality in colleges and universities (ECTQ-GA-BPNN), Artificial intelligence for assessment and feedback to enhance student success in higher education(AIFSHE-ANN)and Sentiment analysis of students’ feedback with NLP and deep learning: A systematic mapping study (SASF-NLP-DL). By comparing other three existing methods, the proposed NLTSEFTQ-MORA-RNN method gives high accuracy respectively.