{"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/learning-and-detecting-concept-drift","title":"Learning and Detecting Concept Drift","arxiv_id":null,"date":"2008-02-01","proceeding":"Graduate School of Information Science and Technology, Hokkaido University 2008 2","authors":["Kyosuke Nishida"],"abstract":"The volume of data that humans create has increased explosively as information\r\nscience and technology have evolved. Therefore, the demand for learning machines that\r\ncan extract input-output mappings and knowledge rules from massive data sets has\r\nbecome more urgent, and machine learning is now a core technology in the advanced\r\ninformation society. It has been applied to fields such as pattern recognition, search\r\nengines, medical support, robot engineering, image processing, and data mining and\r\nhas achieved significant accomplishments in each field. Recently, several methods that\r\ncould not be implemented with older computers have been developed with state-of-the-art computers that have enormous memory capacity and high performance CPUs.\r\nMachine learning is expected to continue to develop in the future.\r\n...\r\nWe propose a simpler method of more accurately detecting concept drift.\r\nThe STEPD method monitors the predictive accuracy of a single online classifier and\r\ndetects significant decreases in the predictive accuracy, which are caused by concept\r\ndrift, by using a statistical test of equal proportions to detect concept drift. When\r\nconcept drift is detected, the online classifier is reinitialized to prepare for the learning\r\nof the next concept. STEPD is able to detect concept drift more quickly and accurately\r\nthan ACE and other methods, but false alarms degrade the predictive accuracy of\r\nSTEPD significantly because STEPD uses only a single online classifier.","url_abs":"https://www.semanticscholar.org/paper/Learning-and-Detecting-Concept-Drift-Nishida/24bbd9e01a9ac44521ac0fe16f77463d2c0849d0?sort=is-influential","url_pdf":"https://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.114.7187&rep=rep1&type=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":"learning-and-detecting-concept-drift","repo_url":"https://github.com/mitre/menelaus","is_official":0,"mentioned_in_paper":0,"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}