Tuesday, September 29, 2015

The evolution of recommendation systems

Great TechCrunch article on the evolution of recommendation systems:

"Just think about how different the experience would be if Netflix didn’t offer future titles to watch, or Amazon didn’t offer alternative products to purchase? What about suggested connections on Facebook and LinkedIn, or news recommendations on Yahoo? None of these would be possible without RS, and that’s just the tip of the iceberg."

1 comment:

  1. Thanks for sharing this article. I have been thinking a lot about recommendation systems recently, and I would like to share some thoughts:

    A) I am always amazed by the ability for more data to beat additional nuances in algorithms. From the Netflix prize, more data from IMDB proved to be very helpful.

    1) http://anand.typepad.com/datawocky/2008/03/more-data-usual.html
    2) http://anand.typepad.com/datawocky/2008/04/data-versus-alg.html

    B) I will be curious to see how willing individuals are to disclose retail preferences in exchange for product discounts or services. Companies such as Stitch Fix do an excellent job of allowing and enticing users to complete surveys to personalize their experience on the platform, and this has been hugely helpful for their model.

    My question for you is which companies could most benefit from allowing users to take surveys to disclose preferences on purchases?