Two months ago, I wrote about Accredible.
I was interested to learn that Accredible proposed that “a living portfolio of evidence that shows you have certain knowledge or skills”. I was reminded of the post today when Stephen Downes linked to news from Fujitsu Laboratories and MIT.
Fujitsu and MIT have announced “a first-of-its-kind, revolutionary asynchronous, personalised learning platform – Guided Learning Pathways”. The platform combines navigation technology with students’ learning behavior simulation based on “an advanced probabilistic learner model”.
Navigation technology will be used to recommend “nuggets” (single, atomistic concepts) for study. Each learner can follow her or his own pathway through the material. The announcement of the partnership between Fujitsu and MIT provided some background to the algorithm approach that will be adopted:
we created a simulation model based on a stochastic, Bayesian Knowledge Tracing algorithm. We also include a topic-graph generator that allows us to generalize across domains and test large-scale systems. Finally, we utilize an implicit rating system for learning materials, in which learning nuggets are not rated by learners directly, and instead their ratings are calculated based on learning outcome of learners. These components’ parameters were set using published field evaluations and give us a general sense of which recommendation algorithms we should test in a deployed system.