A deep-learning-based sensing and evaluation model for teacher instructional competence

Abstract

With the continuous advancement of education informatization, how to scientifically and accurately diagnose and assess the teaching ability of teachers has become an important issue to improve the quality of education. Traditional assessment methods have problems such as high subjectivity and low efficiency, while deep learning technology integrated with behavior sensing and intelligent recognition provides a new solution for teaching ability assessment. In this study, we constructed an intelligent diagnosis and assessment model of teachers' teaching ability based on deep learning, and designed a complete process from multi-source data sensing and collection, feature extraction to model training and validation. Through the processing and analysis of multi-source teaching behavior data, this paper realizes the multidimensional and intelligent assessment of teachers' teaching ability. The experimental results show that the constructed model achieves good performance in terms of assessment accuracy, robustness, and generalization ability. The research results not only help to improve and perfect the index system of students' assessment of teaching, but also have important significance in promoting teachers to improve their teaching behavior and teaching ability. The research results provide theoretical basis and technical support for promoting the intelligent development of education.