Another effective element are the brief self-test quizzes. These one slide multiple choice quizzes occur several times during a lecture and can really snap you into back into focus. They are optional.
The course schedule is well structured with weekly assignments to watch one or more lectures, complete a multiple choice review test, and complete the programming exercise. You are encouraged to retake the review test until you get all the answers right. All review question appear to be randomly drawn from a large pool of questions, with possible answers scrambled.
Some review question appear to be algorithmically generated.
The programming exercises use Octave, which is a free, open-source version of MATLAB. The typical programming exercise calls for completing a small number of Octave functions which fit neatly into some incredible applications. The course provided a submission script for each solution as you work thru the programming exercise. Behind the scenes, some infrastructure automatically tested the solutions and accumulates your score accordingly.
Yet another component is the Q&A Forum which allows to you follow posts of interest.
I used Evernote to screen capture the slides at key moments and to document my programming exercises.
At the end of the course, I received a Statement of Accomplishment.
December 31, 2011
Dear Jan Alden Cornish (janaldencornish@gmail.com),
Congratulations! You have successfully completed the online Machine Learning course (ml-class.org), offered October through December, 2011. To successfully complete this online course, students were required to watch lectures, complete review questions, and work through programming exercises. Your score on these components were as follows:
- Review Questions: 80 out of a maximum of 80
- Programming Exercises: 800 out of a maximum of 800