The New Yorker is America's preeminent weekly magazine for the arts, foreign affairs, and other integral subjects, and the contest draws smart people around the world who think their words deserve to be in the magazine's hallowed pages.
I always thought for New Yorker captions, you read it, there's a microsecond of blankness, and then it hits you and you find it funny. His full response is in McGraw's blog post. Note this is assigned when the page is visited; if the same user visits the page twice, they will get two participant IDs.
In these two experiments, example captions are below: The user is given a URL and after visiting the URL, the user is presented with a series of queries similar to the one above.
Once the submissions have been thinned to fewer than 10, Mankoff sends these semifinalist captions to about 40 other New Yorker editors, who then give him crucial feedback before he chooses the three finalists, which appear in the magazine. Even some winners aren't absolutely convinced of their caption's superiority.
Dataset high-level description These datasets are part of the "cartoon caption contest" where given a cartoon users are supposed to write a funny caption. Captions with uncommon words were more likely to make the shortlist. Mine was: 'Quick -- order a pitcher of water, no ice.
Pellack's caption: "Yes, he's still out there -- and he looks serious. Our algorithms help determine the funniest algorithms and we also provide unbiased data. They're an interesting mix of social commentary, wry New York sensibilities and deadpan punch, among other things.
First, they had to figure out what data they could glean from the captions.