How Khan Academy is using machine learning to assess student mastery
1–10 of 48 posts
Re: How Khan Academy is using machine learning to assess student mastery
#2How soon until machines will be teaching us and grading us?
Re: How Khan Academy is using machine learning to assess student mastery
#3It's interesting how simple their feature set is. I imagine the two EMAs and the percent_correct are probably the most important inputs. A few other interesting ones might be the percent correct for say the past 10 or 20 questions (instead of explicitly cutting off the last 20 questions as they mention). They may also want to pick a different response, like % of next 10 the user gets right instead of probability of getting the next question right.
Re: How Khan Academy is using machine learning to assess student mastery
#4I think we should note that logistic regression has been around forever, and should probably be considered property of "statistics", not "machine learning".
Re: How Khan Academy is using machine learning to assess student mastery
#5Wow, this is literally the exact application of the Stanford Machine Learning class up to this point.
Re: How Khan Academy is using machine learning to assess student mastery
#6Another interesting mechanism is what ChessTempo.com uses. Players and puzzles both have a Elo-like dynamic ranking. When a player beats a puzzle, the player's rank increases and the puzzle's rank decreases. This makes it easy to, over time, present players with puzzles that are of appropriate difficulty.
Re: How Khan Academy is using machine learning to assess student mastery
#7Wow, this is literally the exact application of the Stanford Machine Learning class up to this point.
I was thinking the exact same thing. It was really neat to read about a real world use of what I'm learning in the class.
It makes me wonder if the data set is available as well.
Re: How Khan Academy is using machine learning to assess student mastery
#8Wow, this is literally the exact application of the Stanford Machine Learning class up to this point.
the professor did say people make careers off of what we already know with linear and logistic regression.
Re: How Khan Academy is using machine learning to assess student mastery
#9It takes just one random book pick to prove God, if He's in the mood. (It will look like time travel.)
God says... C:\LoseThos\www.losethos.com\text\WEALTH.TXT
for such a term of years as might give them time to recover, with profit, whatever they should lay not in the further improvement of the land. The expensive vanity of the landlord made him willing to accept of this condition; and hence the origin of long leases.
Even a tenant at will, who pays the full value of the land, is not altogether dependent upon the landlord. The pecuniary advantages which they receive from one another are mutual and equal, and such a tenant will expose neither his life n
-------
You get out of prayer what you put into it. Deja Vu. :-)
Re: How Khan Academy is using machine learning to assess student mastery
#10>This was a fairly large change that we, understandably, only wanted to deploy to a small subset of users. This was facilitated by Bengineer Kamen's GAE/Bingo split-testing framework for App Engine.
I think this method of A/B testing has some faults. I blogged about it A/B testing. Is Khan doing it wrong?http://liveatthewitchtrials.blogspot.com/2011/09/ab-testing-...
and Allen Downey ran some simulations at Repeated tests: how bad can it be?http://allendowney.blogspot.com/2011/10/repeated-tests-how-b...