ANTICIPATING FAILURE: A COMPREHENSIVE ANALYSIS OF ENTREPRENEURSHIP DYNAMIC FACTORS USING MACHINE LEARNING PREDICTIVE MODELS

Abstract

This study explores how machine learning models, such, as Support Vector Machines (SVM) Decision Trees, Logistic Regression, Random Forest, Ensemble Models and Neural Networks can predict the failures of startups. It highlights the impact of bureaucracy, engagement of human resources, financial capacity and mentorship and coaching within organizations on startups operating in the business environment of Morocco. The correlation analysis indicates that traditional methods need to be reevaluated to anticipate challenges faced by organizations due to the lack of established patterns. The research showcases how machine learning provides flexibility and valuable insights beyond correlation analysis. Specifically Random Forest and Ensemble Models emphasize the importance of bureaucracy and human capital in forecasting business success. Variations in rankings highlight connections that stress the need to comprehend all factors. The complexity of bureaucracy is depicted by its role in both facilitating and hindering progress. Human resources play a role in demonstrating their contributions to the organization. This study underscores the significance of capital and financial resources in overcoming obstacles and promoting growth. These discoveries are set to have implications for strategies while deepening our comprehension of business dynamics, within Morocco.