10 thoughts on “Internet Gender Identification”

  1. Haha, 100% Male. Busted!

    Funny implementation, seems to be skewed towards us-sites though.

    1. Joen says:

      It is skewed, yes, but I would call this more of a proof of concept than anything. Imagine this with a google-size database of gender distribution. I’d wager the results would be prette definitive.

  2. Nicole says:

    Oh dear. Apparently there’s a 79% chance that I’m male, based on my browsing habits, and that’s 100% wrong.

    Yay for being an outlier?

  3. Jenny-fa says:

    GODDAMMIT 100% CHANCE FEMALE. I actually dislike most females. Goddammit.

  4. Joen says:

    Guys, (and girls), don’t take it personally. I just heard that my sister is, apparently also, male. And that’s not such a bad thing. We’re not as pretty or elegant, but we can chop wood up real fast.

    More to the point, I think the database of cookie websites this guy has indexed is skewed and a bit old. If this was ever to be used to, say, direct specific book recommendations at Amazon.com, I would reckon Amazon would expand the database hundredfold, and tweak it continuously using a bayesian algohrithm. Then it might guess right.

  5. Levi says:

    Pretty interesting actually. I turned out 100% male and my girlfriend got ~70% female. There’s definitely a lot of potential in this.

  6. Zarah says:

    Well yes, I also turned out to be more male then female. Kind of fifty fiftyish. It makes a lot of sense though.

    1. Joen says:

      Not sure. If anything you should count as two girls these days 😉

  7. Zarah says:

    Haha you’r so right. Maybe we’ll be surprised on the 26th of April.

Comments are closed.