Digital Ability Enhancement Strategy and Effect Evaluation of Vocational Education Teachers Based on Improved BP and KKT Model

Abstract

The improvement of the digital ability of vocational education teachers (VETs) has become a key link in the modernization of education. To enhance the digital competence of VETs, this study applies the back propagation (BP) Neural Network to improve their digital teaching ability and analyzes the tracking effect through a Keyword Knowledge Tracing (KKT) model based on a keyword structure. Specific research has significantly enhanced the digital capabilities of VETs by improving the BP neural network, optimizing the initial weights, introducing momentum factors, and adopting adaptive learning rates. Meanwhile, the KKT model is used to dynamically track the development trajectory of teachers’ digital capabilities, providing precise support for personalized training. In the experiment, the improved BP shows excellent performance in improving digital ability, with accuracy and recall rates of 0.959 and 0.931, significantly higher than other models. In terms of the improvement rate of digital tool usage, the improved BP is 85.2%. The KKT model also performs well in tracking the development of teachers’ digital abilities, with dynamic adaptability scores ranging from 9.2 to 9.6 and tracking ability stability exceeding 94%. This indicates that the improved BP and KKT models can effectively enhance teachers’ digital abilities. The research results have important practical significance for promoting the digital teaching capacity improvement of VETs and provide a scientific basis for educational institutions to formulate targeted teacher training plans.