The Governance Path of Private Higher Education in Guangdong Province Under the Background of the Digital Economy: An Empirical Study Using Random Forest, K-Means Clustering, and PCA

This study explores the governance of private higher education (HE) in Guangdong Province, China, under the dual pressures of the national classified management policy and the rapidly evolving digital economy. As private HEIs are divided into for-profit and non-profit categories, digital transformation introduces new and complex governance challenges related to institutional identity, fiscal sustainability, policy implementation, and stakeholder coordination. Using a dataset of 436 valid survey responses, the study applies the Random Forest algorithm to identify key predictors of policy support and combines K-Means clustering with Principal Component Analysis (PCA) to construct governance profiles of institutions. The results reveal three main types of governance characteristics—advanced, transitional, and constrained—each facing distinct difficulties in areas such as digital readiness, faculty capacity, financial access, and institutional strategy. The Random Forest model highlights the predictive importance of factors like digital teaching capability, attitudes toward financial incentives, and institutional trust in governance frameworks. Based on these insights, the study proposes targeted solutions, including multi-dimensional classification criteria, digital supervision platforms, and collaborative governance mechanisms, to enhance policy effectiveness and institutional adaptability in the digital era.