2020/02/05 by Nabil Hossain, John Krumm, Hossain, Nabil +5
Computer Science · Psychology · Social Sciences · #Artificial Intelligence (cs.AI) #Computation and Language (cs.CL) #Digital Games and Media #FOS: Computer and information sciences #Humor Studies and Applications #Video Analysis and Summarization
paper · pdf · doi:10.48550/arxiv.2002.02031
openalex publication_date 2020/02/05 · openalex created_date 2022/07/26 · openalex updated_date 2026/07/28
Building datasets of creative text, such as humor, is quite challenging. We\nintroduce FunLines, a competitive game where players edit news headlines to\nmake them funny, and where they rate the funniness of headlines edited by\nothers. FunLines makes the humor generation process fun, interactive,\ncollaborative, rewarding and educational, keeping players engaged and providing\nhumor data at a very low cost compared to traditional crowdsourcing approaches.\nFunLines offers useful performance feedback, assisting players in getting\nbetter over time at generating and assessing humor, as our analysis shows. This\nhelps to further increase the quality of the generated dataset. We show the\neffectiveness of this data by training humor classification models that\noutperform a previous benchmark, and we release this dataset to the public.\n