Research Projects

Entrepreneurship encourages growth, wealth, and well-being, benefiting both New India and the global community. However, traditional regression-based methodologies struggle to capture complex nonlinear patterns in entrepreneurship drivers. To address this, a unique method called machine learning is used to examine the influence of social-cognitive, institutional, and demographic characteristics in early-stage entrepreneurship. This research aims to anticipate and characterize total early-stage entrepreneurial activity (TEA) using supervised learning methods. Data will be analyzed using nonlinear classification models (MARS and SVMs) and classification-tree and rule-based models (random forest, AdaBoost). The work falls under supervised learning, as the target variable (TEA) is known. Four algorithms will be selected based on their popularity and applicability in an industrial setting. The anticipated results will help determine the most influential social-cognitive and institutional variables of TEA and the numerous combinations of characteristics that lead to entrepreneurship.