There is very little innovation in online dating. Most dating sites just allow you to post a picture and a profile, and that's about it. A problem with this is that a self-written profiles isn't a very accurate representation of one's self. Also, people often don't KNOW what they want in dating, so keyword searchs are ineffective.
How about a dating site that uses collaborative filtering for finding people you would naturally get along with? Very similar to Amazon's "if you like this book, you'll also probably like this book". But in this case it would be, "if you are best friends with this person, you are likely to be best friends with this person". You can use myspace and facebook as your dataset and perhaps mish-mash them with other datasets that are also good predictors (for instance credit scores, books you like to read, hotornot scores of types of people you tend to find attractive, etc).
Ian Ayres' excellent SuperCrunchers book discusses the statistical process in a bit more detail. But I always thought that if this works for books, why can't it work for online dating?
Anyone interested in programming something like this with me?
I agree
No opinion
I disagree
Add your comment
so, you want a holistic matching system from different sources. Why not work it into a social media idea where you match based on user profiles and interests from facebook, friendfeed etc. there's one site that i stumbled upon that claims it will do something similar:
http://sneak.playwithmatches.com/public/home.aspx#
One person can rarely acquire a large number of reviews, and most reviews will be given by someone leaving the relationship. Ouch. Plus, Mr. Perfect for one lady is usually Mr. Wrong for 99 other ladies. I agree there is a lot of room for improvement... but I am not sure how to aggregate accurate & useful data.
Perhaps if it was set up kind of like a speed dating thing... where you gave your first impressions and that was the basis for the ratings? Where everyone knew ahead of time that there was no commitment and gave their reviews anonymously to the site. I LOATHE online dating and will be tracking this so hopefully we can fix the system. I think you're on the right track to have it user-driven though... E-harm and Match are rip offs!