An AI-Driven Learning Analytics Model for Evaluating Digital Competence of Vocational Teachers

The rapid digital transformation of vocational education requires teachers to possess advanced digital competence to design, implement, and assess technology-enhanced learning. However, current evaluation approaches rely primarily on subjective surveys and manual observations, which limit accuracy, scalability, and real-time feedback. To address this gap, this study proposes an AI-driven learning analytics model for evaluating and predicting the digital competence of vocational teachers. The model integrates multimodal instructional data—including classroom interaction logs, technology-usage behaviors, resource-development traces, and performance indicators—based on the Digital Competence Framework for Vocational Teachers in China. A hybrid machine learning architecture combining feature engineering, attention-based representation learning, and gradient-boosted prediction is developed to classify competence levels across the framework’s eight dimensions. Experimental results demonstrate that the proposed model significantly outperforms traditional assessment methods, achieving high accuracy and strong generalization in competence prediction. The study provides a scalable, data-driven solution for teacher development, offering practical implications for intelligent evaluation, personalized training, and digital-education reform.