MIT hoping to improve MOOC video lectures

mooc-lectureVideo lectures are a fundamental part of your average MOOC.  Despite the use of relatively modern technology however, the pedagogy of the lectures is not really any different to a physical chalk and talk lecture in any campus or classroom around the world.  That sometimes leads educators to question the revolutionary potential for MOOCs to change now only who can learn, but how they go about doing so.

Obviously it’s not quite as simple as that, and even something as straight forward as being able to re-watch the lectures to your hearts content is a noticeable improvement over the one time only nature of your typical lecture.  Nevertheless, that hasn’t stopped researchers from MIT’s Computer Science and AI Lab (CSAIL) from seeking to make things altogether more effective.

They conducted a study involving over 7 million video lecture sessions, consumed by over 100,000 students, to see if they could figure out how to make them better.  Several key findings emerged from the study:

  • Shorter videos are best, with most students tuning out after 6 minutes
  • Informal presentation styles work better than more formal ones
  • Powerpoint slides suck.  Try and make your visuals as lively as possible
  • Faster talkers were seen as more engaging than their slower talking peers
  • Give students enough time to consume complex diagrams
  • Produce bespoke material for MOOCs rather than repurpose existing material for them

These insights then formed the bulk of the CSAIL teams new video product LectureScape.  LectureScape uses the data mined on viewing behaviour to present MOOC lectures in a more dynamic, intuitive and effective method.  It comes with a number of core features:

  • A timeline shows which parts other users have most frequently watched
  • An interactive transcript lets users enter keywords to find relevant segments
  • A mechanism automatically creates word clouds and summaries of individual sections, as well as the whole presentation
  • Content from popular slides automatically appears in the following slide, as users will likely want to refer back to that information

So basically, students can view lectures in a much better manner, skipping around sections as they require.  The group next hope to build a Netflix style video recommendation service so that we can be notified of similar and relevant lectures elsewhere.

They also hopes to implement the tool on a larger scale to quantify its effect on student engagement and performance. The technology can be easily applied to existing MOOC content, as the tool uses machine learning to automatically segment videos based on visual cues.

It’s a pretty cool development.  Check out the video below for an overview of the tool, which will hopefully be finding its way to a MOOC near you shortly.

Related

Facebooktwitterredditpinterestlinkedinmail

One thought on “MIT hoping to improve MOOC video lectures

  1. I'm not a football fan, but I really enjoyed the online coverage of the World Cup. I think that kind of approach would be great for MOOCs

Leave a Reply

Your email address will not be published. Required fields are marked *

Captcha loading...