AI-driven innovation and entrepreneurship education: a K-means clustering approach for Chinese university students

Abstract Research problem Traditional approaches to talent development in Chinese higher education are increasingly misaligned with the evolving demands of an innovation-driven economy. Despite growing national interest, Innovation and Entrepreneurship (IAE) education in Chinese universities remains in its formative stage, with significant challenges in delivering personalized, data-informed instruction to diverse student populations. Methodology This study introduces a K-Means Clustering (KMC) approach to classify university students based on shared entrepreneurial traits and learning needs. Data were collected through structured surveys assessing entrepreneurial mindset, self-efficacy, and engagement in IAE activities. The KMC algorithm grouped students into three distinct clusters corresponding to undergraduate (UG), postgraduate (PG), and expert-level learners. Results The clustering analysis revealed discrepancies in instructional support across student levels. Specifically, 33% of expert-level students, 41% of PG-level students, and 40% of UG-level students reported insufficient monitoring and guidance in their IAE learning experiences. These findings highlight the need for differentiated pedagogical strategies aligned with student readiness and capabilities. Implications By applying unsupervised machine learning, the study demonstrates how AI-driven analytics can enhance the personalization and effectiveness of IAE education. The results support the integration of clustering models into educational planning, allowing universities to optimize curriculum delivery and better prepare students for participation in innovation-oriented economies.