it’s been only three weeks since the launch of Q&A site Hunch, but already it’s amassed answers to 4.3 million questions. By taking visitors through a series of questions to get to know them better, Hunch can begin to make guesses about the right kinds of answers for each individual by making correlations across between people who answer similarly across different topics. It’s sort of like collaborative filtering for information.
Co-founder Caterina Fake of Flickr fame shares some of the correlations Hunch has been able to find so far. Some of her findings:
* People who believe that alien abductions are real are more likely to blame Nancy Pelosi for the financial crisis.
* One of the best predictors of whether people agree they should switch to a Mac: whether they like to dance. Are PC users less fun? The data has spoken.
* Fake ID users are more likely to be happy at startups.
Also, people who like sports video games are more likely to have had a broken leg at one point (which would explain why they like to sit on the couch and play football instead of go out and, you know, actually play football).
And on the subject of Facebook versus Twitter, the Hunch data suggests that “Facebook people” are more social than Twitterers in actual social situations like a party. Facebook people mingle with more people, including strangers, whereas Twitterers stick to people they know. Other differences? Facebook people spend more on shampoo and like to drink more. Whereas “Twitterers report that they have oily skin.” In other words, Twitter is for geeks. I knew it! Don’t tell Oprah
Hunch’s data set will become available to researchers next week through a special “researchers API.” Can’t wait to find out what other correlations come out of it.
(Photo by Ira Mejías)