The children's machine: rethinking school in the age of the computer
The children's machine: rethinking school in the age of the computer
Turtles, termites, and traffic jams: explorations in massively parallel microworlds
Turtles, termites, and traffic jams: explorations in massively parallel microworlds
Communications of the ACM
Probabilistic Student Modelling to Improve Exploratory Behaviour
User Modeling and User-Adapted Interaction
International Journal of Artificial Intelligence in Education - "Caring for the Learner" in honour of John Self
Poisson-based inference for perturbation models in adaptive spelling training
International Journal of Artificial Intelligence in Education
Hi-index | 0.00 |
The principle of evolving educational content - stepwise and stochastic refinement of educational problems on the basis of a student's previous success - is extended and evaluated for the domain of spelling. Experimental results, generated by an implementation in a classroom setting, show that the principle of evolving educational content is not only sensitive to an individual student's ability but it also allows the student to take novel paths through an open-ended problem space. Learning outcomes are confirmed both quantitatively and qualitatively, and indicate that the technique transfers to a pedagogically challenging domain. We present how a problem space can be generated to effectively and automatically encompass a wider range of educational domains with little designer intervention.