ML-Based Teaching Systems: A Conceptual Framework

What is it about?

Imagine this: You play chess games against your best friend, and you get beaten. Every. Single. Time. Internet chess platforms like Lichess or chess.com provide a machine learning-based teaching system that can guide you through your lost games, identify mistakes, and teach you the optimal next move. By using pre-trained machine-learning models and adapting their output to your match history and learning progress, this system can help you establish a better understanding of good and bad decisions, ultimately leading to victory against your friend. The example above illustrates how machine learning-based teaching systems can enhance the learning process of inexperienced individuals (novices), without the need for intervention from subject matter experts (SMEs). This not only applies to the realm of board games but also demonstrates how such systems can support organizations. To ensure long-term success, it is crucial to provide ongoing employee training, retain the knowledge of retiring experts, and pass it down to newcomers. In the paper, the authors present a framework on ML-based teaching systems to preserve and convey expert knowledge within organizations.

Featured Image

Photo by Arseny Togulev on Unsplash

Photo by Arseny Togulev on Unsplash

Why is it important?

Our findings reveal how ML is used in teaching systems. Furthermore, we conceptualize constructs of ML-based teaching systems in a framework and reveal a future agenda for research.