I'm a bit ambivalent about that feature. I do think many folks will want/expect it, and it would be useful at times. At the same time, the primary use of the app I personally wanted is to add things I wanted to remember permanently, and let the algorithms handle all of the work of scheduling optimal practice. Gonna think about this.

Thanks for the feedback.

Bulk import would be a pretty fast and simple add; I'll make a GitHub issue for it.

Not currently, though that would be quite cool. For now, I clarified my version of that card, which unfortunately had a typo. Thanks..

This is not naive bayes and does not assume independent observations on the exercises. The point of using a network is to model the joint distribution with dependencies.

The 'E' in the regression is the inferred/predicted value of the E variable for that exercise, using no problem history from that exercise--only what's pulled in through the Bayes net. (Sorry that wasn't clear)

The 'T' variable is likely just a case of multicollinearity with the 'E' variable and should go away on a full-scale data set. If not it can easily be removed from the model. The 'E' variable is dominating because is additionally captures cross-sectional affects across the various exercises in the regression.

Ah, ok. So is this a sort of Markov model, where you are predicting the probability of getting an exercise right after observing (some subset of) the previous exercises? And E is not 1 or 0, but the expected probability of getting it right? I'm still confused where all the different E_i's fit in.

That would explain the magnitude, and I agree the negative weight on T would just be due to the direct correlation between E and T.

Edit: I just realized that an exercise consists of multiple problems, so you're predicting whether or not the student will get >= 85% of the problems right on an exercise.