Smart Entrepreneurial Systems: An Application of Deep Reinforcement Learning in Improving Entrepreneurship Mentorship

It is imperative to help entrepreneurs learn creative and strategic approaches to enterprise; however, there are limited efforts to design effective educational platforms, in particular mentorship mechanisms, for novice entrepreneurs due to limitations such as scalability, inclusivity and customizability. Addressing this gap, this study proposes a new analytical solution—the development of a smart mentorship agent (a mentor bot) for entrepreneurship education platforms. This smart agent, if developed, will be able to support novice entrepreneurs by offering them accessible, modular, personalized, and adaptive learning experiences in entrepreneurship. The proposed agent uses reinforcement learning (RL) algorithms to learn critical success factors for learning content recommendation. The BizzyB platform, an online entrepreneurship education platform, was used as a testbed to demonstrate the applicability and utility of the proposed algorithm. Using multi-layer Deep Q-learning networks, this study has theoretical and practical implications by offering a new approach to developing smart entrepreneurial systems, validating their effectiveness, and supporting data-driven entrepreneurship education.