https://github.com/ddofer/cah
Code used for Cards Against Humanity EMNLP paper
https://github.com/ddofer/cah
cards-against-humanity humor-classification machine-learning nlp
Last synced: 12 months ago
JSON representation
Code used for Cards Against Humanity EMNLP paper
- Host: GitHub
- URL: https://github.com/ddofer/cah
- Owner: ddofer
- Created: 2022-06-24T10:09:31.000Z (about 4 years ago)
- Default Branch: main
- Last Pushed: 2023-09-04T13:20:19.000Z (almost 3 years ago)
- Last Synced: 2025-07-08T23:40:01.653Z (12 months ago)
- Topics: cards-against-humanity, humor-classification, machine-learning, nlp
- Language: Jupyter Notebook
- Homepage:
- Size: 1.25 MB
- Stars: 11
- Watchers: 1
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# CAH
## `It’s All in the Cards: Predicting Humor in a Fill-in-the-blank Party Game`
Dan Ofer and Dafna Shahaf
Paper: https://arxiv.org/abs/2210.13016
Repo contains code and results used for Cards Against Humanity NLP paper, published at [EMNLP 2022 findings](https://preview.aclanthology.org/emnlp-22-ingestion/2022.findings-emnlp.394.pdf).
Data available upon request from CAH labs: [https://lab.cardsagainsthumanity.com/](https://lab.cardsagainsthumanity.com/)
----------------------------------------------------------------------------------------------------
```
@misc{https://doi.org/10.48550/arxiv.2210.13016,
doi = {10.48550/ARXIV.2210.13016},
url = {https://arxiv.org/abs/2210.13016},
author = {Ofer, Dan and Shahaf, Dafna},
abstract = {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.}
keywords = {Machine Learning (cs.LG), Artificial Intelligence (cs.AI), Computation and Language (cs.CL), Computers and Society (cs.CY), General Literature (cs.GL), FOS: Computer and information sciences, FOS: Computer and information sciences, I.2.7; I.2; K.4; J.4; J.5, 68T01, 68T50},
title = {Cards Against AI: Predicting Humor in a Fill-in-the-blank Party Game},
publisher = {arXiv},
year = {2022},
copyright = {Creative Commons Attribution Non Commercial No Derivatives 4.0 International}
}
```