My question is how does time dependency work in this case. I am trying to wrap my head around how a prediction engine would work when your assessing students on the basis of not just past/current performance but also how much time their taking while answering each question.
I think you can model for randomness (kids getting lucky while answering a question), but if you can somehow add time-dependency to the model, then your predictability would be higher (of course this is pure speculation).
Does anyone have a good model I can look at? Any help would be appreciated.
I think you can model for randomness (kids getting lucky while answering a question), but if you can somehow add time-dependency to the model, then your predictability would be higher (of course this is pure speculation).
Does anyone have a good model I can look at? Any help would be appreciated.