머신러닝을 활용한 지식재산기반 스타트업 최고경영자 핵심역량 도출에 관한 연구

초록

This study aims to identify the core competencies of CEOs, which are crucial for the performance of IP-based startups, reflecting their increasing importance. To date, no research has been conducted, either domestically or internationally, analyzing the core competencies of CEOs in IP-based startups. Despite the importance of CEOs, the competencies that form their foundation have not been systematically identified. In this study, we propose a innovative competency modeling method by mining data based on the opinions of actual CEOs from IP-based startups and applying machine learning. The machine learning-based competency modeling method is significant in that it expands the scope from the traditional event-based identification methods to those based on prediction and accuracy. Given the characteristics of IP-based startups, which must adapt to rapidly changing business environments and technologies, this method overcomes the limitations of traditional competency modeling, allowing for more responsive strategies. In our analysis, we applied machine learning algorithms such as Random Forest, Bayesian Network, and Gradient Boosting. Notably, we employed a backward elimination method to derive core competencies that have a close impact on actual high-growth indicators and ensured high prediction rates and accuracy through various ensemble techniques. The 21 core competencies identified through this study include information on their importance and prediction ranking. These insights can be utilized in various fields such as direction setting, evaluation, and nurturing, not only for current CEOs of IP-based startups but also for prospective entrepreneurs.

키워드