Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Here's an example using our tool. You can see that Lost must not have aired this Wednesday by looking at the chart. Also, people discuss Lost 3x as much as Heroes:

http://twist.flaptor.com/trends?gram=lost,heroes&table=1

Another interesting pattern is Lost vs. American Idol:

http://twist.flaptor.com/trends?gram=lost%2Cidol&table=1



Yeah that is just goes to show the demographic on Twitter now. Not of the mainstream. The MySpacers are now just moving to Facebook, but if the mainstream was on Twitter, that chart would favor American Idol.


So how would you separate "lost wast good last night" from "i cant believe uconn lost"?


Because the volume of tweets about Lost when it airs is so high that it trumps everything else. Here is perhaps a better example:

http://twist.flaptor.com/freq?gram=american&table=0

If you click on the peaks (caused by American Idol) you'll see the results for a query on the term "American" at that time. A significant number of those mention American Idol.

Now click near the right of the chart. Almost no results mention the show.


Fair enough. Keep up the good work.


some of those "lost" example tweets are 'lost' as in 'lost my job' :(




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: