Are PPP projects poor fundability in China：Based on ensemble learning

Enhancing the participation of Public-Private Partnership (PPP) social capital crucial for stimulating effective investment. Given the myriad of factors influencing social capital's participation in PPP projects, this study examined 14,038 PPP projects from the China Public-Private Partnership Center (CPPPC) database. We aim to construct an ensemble learning-based prediction model for PPP project social capital participation by selecting 61 feature variables across four dimensions: project characteristics, local government, market environment, and macroeconomics. Experimental results demonstrate that the dragonfly algorithm effectively improves model prediction accuracy by reducing feature dimensionality. From a feature combination perspective, combinations related to the project itself, local government, and macroeconomic factors exhibit superior predictive performance compared to other combinations. Among these, the project's intrinsic characteristics have the most significant impact on social capital participation in PPPs, followed by local government factors. Notably, the inclusion of market environment variables tends to decrease model accuracy, except when considering the project's intrinsic characteristics alone. Consequently, a dynamic approach is recommended to capture instantaneous variables like the market environment, thus enhancing model performance. Additionally, social capital should assess the financial capacity of local governments to ensure effective PPP outcomes.