The art of predictive analytics: More data, same models

  • February 2, 2012 · 7:00 PM
  • This location is shown only to members

We're really excited for our next event with Joseph Turian!  Come at 6:30 to network with the group.


ABSTRACT:This talk will present a handful of recipes for cheaply improving the accuracy of existing models.


Sure, if you want more accuracy, you can read the latest papers and spend a bunch of time implementing new fangled models. It's definitely fun to brag about how you took the long road. But what if you don't have the time to invest in something nebulous and tricky, and simply want to solve your problem?

A simpler approach is to take your existing model, and shove more data in. You can annotate more data, with active learning. You can insert features from an unsupervised model. You can use distant supervision.  By presenting a handful of recipes and examples, this talk will encourage you to think outside the box, and figure out how to use more data to improve existing models.

This talk is the opposite of my O'Reilly Strata Conference talk "New Developments in Large Data Techniques". That talk was about awesome new models that no one in industry is using yet:http://www.youtube.com/watch?v=fEUw8igr1IY

 
Joseph Turian, Ph.D., is the founder of MetaOptimize, a consultancy onML and NLP.

Join or login to comment.

  • Alex Lin

    Joseph's study slides: http://files.meetup.com/1542972...­

    February 14, 2012

  • Gary Kazantsev

    Waste of time.

    February 4, 2012

  • Kilian

    Joseph is very knowledgeable, and it was an enjoyable talk. A bit rushed but certainly entertaining and useful.

    February 3, 2012

  • Brian Dalessandro

    Are there any follow up materials to post?

    February 3, 2012

  • Yin Aphinyanaphongs

    I compiled a list of references from Joseph's talk. It is far from complete. If you took notes, please add references I missed. https://docs.google.com/document...­

    February 3, 2012

  • A former member
    A former member

    The content of this particular meetup was professional and interesting. Delivery was a little fast, though, making it hard to keep up or take notes. But the speaker was plainly excited about his topic.

    February 3, 2012

  • Jimmy Shen

    I thought that this meetup was very informative in the perspective of a novice in predictive analytics. There could have been more introductory information but I dont believe that this was the target audience of the speaker or of this meetup group.

    February 3, 2012

  • Gustavo Corral

    The content of the talk is interesting, but maybe each of the parts needs further exploration.
    Also, anyone else think 1960s 'geek' style is out ?

    February 3, 2012

  • Ryan Raaum

    Thought provoking. A nice whirlwind tour of tricks and tips.

    February 3, 2012

  • max khesin

    I love cheap tricks to get more data! They are way undervalued

    February 3, 2012

  • Barney Tam

    Very informative.

    February 3, 2012

  • max khesin

    Some people asked about Stanford PGM course videos - here we go http://openclassroom.stanford.e...­

    February 2, 2012

People in this
Meetup are also in:

Sometimes the best Meetup Group is the one you start

Log in

Not registered with us yet?

Sign up

Meetup members, Log in

By clicking "Sign up" or "Sign up using Facebook", you confirm that you accept our Terms of Service & Privacy Policy