Explainable Artificial Intelligence in education
This paper introduces "a framework, referred to as XAI-ED, that considers six key aspects in relation to explainability for studying, designing and developing educational AI tools" where explainability is presented in terms of fairness, accountability, transparency, and ethics (FATE). It then illustrates the application of XAI-ED in four case studies of adaptive educational systems (AESs) that use data about students and learning processes. The framework is fairly straightforward, describing several facets of each of the six aspects, but the real value of the paper is in the case studies, where we get a good look at these four applications: RiPPLE, FUMA, AcaWriter and TeamWork Analytcis.
Today: 2 Total: 460