Cards Against AI: Predicting Humor in a Fill-in-the-blank Party Game
This work addresses the NLP challenge of understanding humor for applications in human-computer interaction, but it is incremental as it applies existing methods to a new dataset.
The paper tackled the problem of predicting humor in the context of Cards Against Humanity by introducing a dataset of 300,000 online games and training machine learning models to predict winning jokes, achieving performance twice as good as random (20%).
Humor is an inherently social phenomenon, with humorous utterances shaped by what is socially and culturally accepted. Understanding humor is an important NLP challenge, with many applications to human-computer interactions. In this work we explore humor in the context of Cards Against Humanity -- a party game where players complete fill-in-the-blank statements using cards that can be offensive or politically incorrect. We introduce a novel dataset of 300,000 online games of Cards Against Humanity, including 785K unique jokes, analyze it and provide insights. We trained machine learning models to predict the winning joke per game, achieving performance twice as good (20\%) as random, even without any user information. On the more difficult task of judging novel cards, we see the models' ability to generalize is moderate. Interestingly, we find that our models are primarily focused on punchline card, with the context having little impact. Analyzing feature importance, we observe that short, crude, juvenile punchlines tend to win.