Hybrid Models of Machine learning based Multi-Agent Coordination for Optimized Course Selection in E-Learning Environment

In the sphere of e-learning, the optimization of course selection strategies poses a critical challenge due to the varied preferences and ever-evolving nature of learners. This study explores how multi-agent coalition building might improve the e-learning environment's course selection process. The primary focus lies in devising collaborative strategies where multiple agents, representing students or learners, collaborate through coalition formation to allocate courses efficiently while respecting the unique preferences of individual learners. This study revolves around the formation of coalitions among agents tasked with selecting courses for university. Introducing a pioneering voting approach, agents form coalitions and allocate points to different courses through several rounds of voting. This method allows agents to express their preferences freely while intelligently leveraging information from previous rounds to vote strategically. Furthermore, different voting schemes for the coalition method course selection in the university are presented and evaluated in detail in a variety of scenarios. The findings reveal that even a simplistic voting strategy yields outcomes closely aligned with optimal results. Interestingly, the more sophisticated strategy failed to exploit other, less savvy voters, indicating potential equity within the e-learning domain—an encouraging prospect for institutions striving for fairness and impartiality.