The Evaluation Model of Physical Education Teaching Performance based on Deep Learning Algorithm

Abstract

The various manifestations of physical education instructional elements and assessment ambiguity affect both the qualitative and quantitative evaluation outcomes of instructional effects. An evaluation approach of physical education teaching and training excellence based on deep learning is developed to address the issues of high complexity and low accuracy in the assessment of physical education teaching outcomes. The construction of the system of evaluation indexes is predicated on the instructional material, teaching behaviour, instructional resources, teaching technique, and learning impacts that impact the quality of instruction. The model for the assessment of the physical education learning effect was created using the Genetic Algorithm-Back Propagation Neural Network (GA-BPNN) to increase the assessment accuracy of the teaching effect. The monitoring of the entire teaching process is the foundation of the assessment approach. The general objectives and a chosen set of three-level indicators for evaluation were examined considering the different types of physical education teaching variables and the assessment’s degree of uncertainties. The hierarchical structure was created using the (GA-BPNN) method, which also produced the hierarchy’s overall rating and comprehensive score. According to the test results, the teaching assessment model developed in this work has an Accuracy, which is higher than the industry average and supports improving the quality of instruction.