讲座简介：
Enhancing Startup Financing Prediction by Integrating All Venture Capitalists in the Network: A Machine Learning Approach
Abstract
This study employs a machine learning (ML) algorithm, specifically heterogeneous graph neural network (HGNN), to incorporate startups’ connections with all VCs—not just lead VCs—in predicting startup financing. Drawing on network substitution perspective, we develop a hybrid composition layer to aggregate diverse networks (i.e., investment, syndication, and alliance networks). We find that including all VCs substantially enhances prediction accuracy. Moreover, applying network substitution perspective to ML further improves accuracy. Additionally, we compare the ML results with those from regression models and observe a parallel trend in model improvement, which enhances the explainability of the ML approach. Our study contributes by demonstrating how and why blending human insights, such as theoretical frameworks, with the strengths of ML algorithms can refine strategic decision-making for VC investors.
主讲人简介：
Dr. Jin Chen is an Associate Professor of Entrepreneurship and Innovation at the Nottingham University Business School China, University of Nottingham Ningbo China (UNNC). She received her PhD from the National University of Singapore. Prior to joining UNNC in 2018, she was an Associate Professor at the East China University of Science and Technology. Her research focuses on public-private interaction in the context of innovation and entrepreneurship, especially the impact of government venture capitalists and government subsidies on the growth of high-tech start-ups. Her publications have appeared in many reputable journals, including Research Policy, Strategic Entrepreneurship Journal, and the Quarterly Journal of Management (Chinese), among others.
Enhancing Startup Financing Prediction by Integrating All Venture Capitalists in the Network: A Machine Learning Approach
Abstract
This study employs a machine learning (ML) algorithm, specifically heterogeneous graph neural network (HGNN), to incorporate startups’ connections with all VCs—not just lead VCs—in predicting startup financing. Drawing on network substitution perspective, we develop a hybrid composition layer to aggregate diverse networks (i.e., investment, syndication, and alliance networks). We find that including all VCs substantially enhances prediction accuracy. Moreover, applying network substitution perspective to ML further improves accuracy. Additionally, we compare the ML results with those from regression models and observe a parallel trend in model improvement, which enhances the explainability of the ML approach. Our study contributes by demonstrating how and why blending human insights, such as theoretical frameworks, with the strengths of ML algorithms can refine strategic decision-making for VC investors.
主讲人简介：
Dr. Jin Chen is an Associate Professor of Entrepreneurship and Innovation at the Nottingham University Business School China, University of Nottingham Ningbo China (UNNC). She received her PhD from the National University of Singapore. Prior to joining UNNC in 2018, she was an Associate Professor at the East China University of Science and Technology. Her research focuses on public-private interaction in the context of innovation and entrepreneurship, especially the impact of government venture capitalists and government subsidies on the growth of high-tech start-ups. Her publications have appeared in many reputable journals, including Research Policy, Strategic Entrepreneurship Journal, and the Quarterly Journal of Management (Chinese), among others.
沪公网安备 31011502006855号