An English teaching quality evaluation model based on Gaussian process machine learning

Shi Qi,Lei Liu,B. Santhosh Kumar,A. Prathik

DOI: https://doi.org/10.1111/exsy.12861

IF: 3.3

2021-12-24

Expert Systems

Abstract:The efficiency of conventional English teaching quality evaluation is comparatively small, and evaluation statistics are challenging. To investigate the use of artificial intelligence (AI) technology in teacher teaching assessment, a machine learning algorithm is proposed to create a teaching evaluation model suitable for the current educational model to assist colleges and universities in overcoming existing teaching challenges. The proposed Machine learning‐based Gaussian process model (MLGPM) improves the student's language skills. The proposed model uses Gaussian mixed model to express the circulation features of samples and enhances the support vector machine. Therefore, this paper suggests an active learning algorithm that, in association with Gaussian mixed model and sparse Bayesian learning, strategically chooses and labels samples to construct a classifier that syndicates the distribution characteristics of the samples. As a result, the accuracy of a considered quality index for English classrooms is verified, and the quality and control of English as a foreign language can be enhanced. The experiment results show that the model presented in this study is effective and beneficial when assessing the efficiency of teaching in universities and analyzing big data sets. The simulation analysis with student performance improvement in English teaching quality using machine learning high fluency rate of 95.3, high accuracy ratio of 98.1%, improve vocabulary prediction ratio of 94.6%, improve passage prediction ratio of 92.7%, enhance learning rate of 95.2%, reduce the error rate of 24.1%, F1‐score of 91.5% and assessment score of 92.1% when compared with other methods.

computer science, artificial intelligence, theory & methods