Leveraging generative AI, knowledge graphs and personalised learning paths to improve the learning experience of neurodivergent students in e-learning courses

Around 20% of the world’s population is affected by neurodiversity, often leading to increased difficulties when operating in an educational setting designed for neurotypicals. There is a strong need for effective, tailored education to help reduce the friction between the different lived experiences of members of affected groups. Although e-learning systems, especially those with personalised learning paths, provide significant support in overcoming the limitations of teaching strategies, they typically do not offer sufficient customisation options to meet the needs of neurodivergent students. While e-learning systems offer assistive tools, they lack empathy and adaptive information structuring, both of which are essential for providing adequate support to neurodivergent individuals and enhancing their learning experience. We propose a framework for personalised learning paths based on knowledge graphs, incorporating an LLM solution that can customise content for neurodivergent learners, helping them improve their knowledge and skills. By leveraging matchmaking software that evaluates content quality using empathy, creativity and tone sensitivity metrics, the framework can effectively assess responses and alignment to address the needs of neurodivergent individuals.

This version of the paper has been accepted for publication, after peer review and is subject to Springer Nature’s AM terms of use (https://www.springernature.com/gp/open-science/policies/accepted-man...), but is not the Version of Record and does not reflect post-acceptance improvements, or any corrections. The Version of Record is available online at: https://doi.org/10.1007/978-3-032-23260-1_20.