ġ6 5 if that ’s my cow, tell her i left for pamplona. ġ5 5 if it ’s my wife, i was working late on a merrill-lynch commercial. ġ4 5 if that ’s merrill lynch, tell them i ’ll be back when i ’m good and ready. ġ2 4 jeez ! what ’s a minotaur got to do to get a drink around here ?ġ3 4 if i hear that ’ a guy and a minotaur go into a bar ’ joke one more time. ġ1 4 if that ’s my wife tell her i ’ll be home in a minotaur. ġ0 4 if it ’s my wife, tell her that i ran into an old minotaur friend. ĩ 5 if that ’s my wife, tell her i went to pamplona. Ĩ 5 if that ’s merrill lynch, tell them i quit and went to pamplona. ĥ 2 if that ’s bessie, tell her i ’ve moooooved on !Ħ 3 if its my wife, tell her i ’m in a china shop. Ĥ 1 if that ’s elsie, you have n’t seen me. ģ 0 if it ’s theseus, tell him to get lost. ġ 0 if it ’s theseus, tell him i ’ll be back in the labyrinth just as soon as happy hour is over. The tfidf scores used to build the lexical network are used in method OR4. , to cluster the captions in each contest thematically the sizes of these clusters comprise method OR3. We also used a graph clustering method, previously used in King et al. To identify the most central caption in each contest (method OR1 ) and the one with the highest lexrank score (method OR2 ). We built a lexical network out of the captions for each contest. Figure 1: Cartoon number 31 Figure 2: Cartoon number 32 2 Related Work Then, we performed Amazon Mechanical Turk experiments in which we asked Turkers to judge which of the selected captions is funnier. We used each of these methods to independently rank all captions from our corpus and selected the top captions for each method. We developed a set of unsupervised methods for ranking captions based on features such as originality, centrality, sentiment, concreteness, grammaticality, human-centeredness, etc. In this paper, we take a computational approach to studying the contest to gain insights into what differentiates funny captions from the rest. The contest has become a cultural phenomenon and has generated a lot of discussion as to what makes a cartoon funny (at least, to the readers of the New Yorker). They pick the top 3 submitted captions and ask the readers to pick the weekly winner. Figures 1 and 2 ) and ask readers to come up with a funny caption for it. Each week, the editors post a cartoon (cf. The New Yorker Cartoon Caption Contest has been running for more than 10 years.
0 Comments
Leave a Reply. |