Profiling (Non-)Nascent Entrepreneurs in Hungary Based on Machine Learning Approaches

In our study, we examined the characteristics of nascent entrepreneurs using 2021 Global Entrepreneurship Monitor national representative data in Hungary. We topic based on Arenius and Minitti’s four-category theory framework. research, system-level feature sets with four machine learning modeling algorithms: multivariate adaptive regression spline (MARS), support vector (SVM), random forest (RF), AdaBoost. Our results show that each algorithm can predict over 90% cruise control (ACC) accuracy. Furthermore, adaptation categories variables Minitti provides an appropriate framework for obtaining reliable predictions. Based results, it be concluded perceptual factors have different importance weight along optimal models, if include further reliability measures model validation, cannot pinpoint only one adequately identify entrepreneurs. Accurate forecasting requires a careful predictor-level analysis algorithms’ which also includes systemic relationship between affecting factors. An important but unexpected result study is identified Hungarian NEs very specific previous entrepreneurial business ownership experience; thus, they defined not as beginner novice enterprise.