Explainable artificial intelligence driven dynamic evaluation framework for Industry 4.0 oriented talent cultivation in higher education

Abstract

Abstract The rapid advancement of Industry 4.0 technologies has created an urgent demand for highly skilled graduates equipped with competencies in artificial intelligence, data analytics, Internet of Things (IoT), cloud computing, cybersecurity, digital manufacturing, and innovation management. To address this challenge, this research proposes an Explainable AI (XAI)-Driven Dynamic Evaluation Framework for assessing and optimizing Industry 4.0 talent cultivation in higher education institutions. The proposed model utilizes the College Student Placement Factors Dataset containing 10,000 student records with attributes such as academic performance, communication skills, internships, and project experience, to evaluate talent cultivation outcomes. Min–Max normalization is applied to standardize heterogeneous educational attributes, ensuring data consistency and balanced feature contribution during model training. Kernel Principal Component Analysis (KPCA) is employed to extract informative nonlinear competency representations, reducing feature redundancy and enhancing predictive learning capability. Subsequently, a hybrid deep learning model combining Chaotic Dragonfly Algorithm-tuned Attention-Based Long Short-Term Memory (CDA-Attention-LSTM) is developed to dynamically evaluate students' Industry 4.0 competency levels and predict talent cultivation effectiveness. To enhance transparency and trustworthiness, SHAP (Shapley Additive Explanations) is used to identify key factors influencing evaluation results and provide interpretable recommendations for educators and administrators. Experimental evaluation is conducted using Python-based tools. Results demonstrate that the proposed model achieves superior prediction accuracy (98.92%), strong evaluation reliability, and high adaptability compared with conventional assessment approaches. The proposed model offers an effective decision-support tool for curriculum optimization, educational quality enhancement, and sustainable Industry 4.0 workforce development, thereby strengthening the alignment between higher education outcomes and evolving industrial requirements.

The authors' abstract, as published at the source. Discover Artificial Intelligence, 2026 · DOI ↗