From data to recommendations with Hunch

While reading September issue of WIRED, I found an interesting article about recommendation. Caterina Fake, also co-founder of Flickr, is chief product officer of Hunch, a social search engine, which goal is to personalize the internet. For this, Hunch build a profile of your tastes and preferences and use these to make recommendations.

Here is an excerpt of the “About” section of Hunch:

After getting to know you by asking you some fun and insightful questions, Hunch will offer you a great recommendation to address your choice, problem, or dilemma, on tens of thousands of topics. Hunch’s recommendations are based on the collective knowledge of the entire Hunch community, narrowed down to people like you, or just enough like you that you might be mistaken for each other in a dark room. Hunch is designed so that every time it’s used, it learns something new. That means Hunch’s hunches are always getting better.

graph-bg

I found the article interesting since at FinScore, we also mine behavioral data plus offline data to make recommendations. These recommendations are used to personalize user experience (content, product, news, ads, etc.).

For more information, read the full article

Share

Recommended Reading

Comments Icon3 comments found on “From data to recommendations with Hunch

  1. I’ve played around with Hunch and it’s a lot of fun.

    The recommendation algorithm is still not up to my hopes though. I wish for a recommendation engine that takes my last.fm playlist and tells me what jobs I will like. OK, maybe it needs a little bit of related info … but I wish it would make more leaps of judgment.

Comments are closed.