Conducting Rapid Experimentation With An Open-Source Adaptive Tutoring System

Intelligent Tutoring Systems have been an area of particular relevance and importance to AIED. In this tutorial, we introduce uses of a new tool to accelerate the speed at which the community can innovate and experiment in the general area of computer tutoring systems. We showcase the field’s first fully fledged and open-source adaptive tutoring system (OATutor) with a completely creative commons licensed problem library based on popular open-licensed algebra textbooks. We demonstrate, with hands-on tutorials, how the system can be deployed on github.io and used to help AIED researchers rapidly run A/B experiments, including how to add or modify content in the system, analyze its log data, and link OATutor content to learning management system with LTI. Our open-sourcing of three textbooks worth of questions and tutoring content in a structured data format (JSON) also opens up avenues for AIED researchers to apply new and legacy educational data mining and NLP techniques. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.