{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/on-cognitive-preferences-and-the-plausibility","title":"On Cognitive Preferences and the Plausibility of Rule-based Models","arxiv_id":"1803.01316","date":"2018-03-04","proceeding":null,"authors":["Johannes Fürnkranz","Tomáš Kliegr","Heiko Paulheim"],"abstract":"It is conventional wisdom in machine learning and data mining that logical\nmodels such as rule sets are more interpretable than other models, and that\namong such rule-based models, simpler models are more interpretable than more\ncomplex ones. In this position paper, we question this latter assumption by\nfocusing on one particular aspect of interpretability, namely the plausibility\nof models. Roughly speaking, we equate the plausibility of a model with the\nlikeliness that a user accepts it as an explanation for a prediction. In\nparticular, we argue that, all other things being equal, longer explanations\nmay be more convincing than shorter ones, and that the predominant bias for\nshorter models, which is typically necessary for learning powerful\ndiscriminative models, may not be suitable when it comes to user acceptance of\nthe learned models. To that end, we first recapitulate evidence for and against\nthis postulate, and then report the results of an evaluation in a\ncrowd-sourcing study based on about 3.000 judgments. The results do not reveal\na strong preference for simple rules, whereas we can observe a weak preference\nfor longer rules in some domains. We then relate these results to well-known\ncognitive biases such as the conjunction fallacy, the representative heuristic,\nor the recogition heuristic, and investigate their relation to rule length and\nplausibility.","url_abs":"http://arxiv.org/abs/1803.01316v4","url_pdf":"http://arxiv.org/pdf/1803.01316v4.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"on-cognitive-preferences-and-the-plausibility","repo_url":"https://github.com/kliegr/rule-length-project","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}