Predikce výše equity investic do start-upů metodami strojového učení

This thesis investigates the application of tree-based algorithms to analyze factors influencing startup success and evaluates their accuracy. The thesis employs three types of algorithms: Random Forest, Gradient Boosting, and simple regression tree. By applying these algorithms to a startup dataset, the research aims to determine their suitability for analyzing factors that may impact a startup´s success, defined as the total funding raised by a company in U.S. dollars. The findings indicate that tree-based algorithms can serve as practical supplementary tools for decision-making processes by identifying key factors contributing to startup success. However, the study also highlights certain limitations of these methods, which are discussed in detail to provide a balanced perspective on their application.