{"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/concept-drift-detection-for-streaming-data","title":"Concept Drift Detection for Streaming Data","arxiv_id":"1504.01044","date":"2015-04-04","proceeding":null,"authors":["Heng Wang","Zubin Abraham"],"abstract":"Common statistical prediction models often require and assume stationarity in\nthe data. However, in many practical applications, changes in the relationship\nof the response and predictor variables are regularly observed over time,\nresulting in the deterioration of the predictive performance of these models.\nThis paper presents Linear Four Rates (LFR), a framework for detecting these\nconcept drifts and subsequently identifying the data points that belong to the\nnew concept (for relearning the model). Unlike conventional concept drift\ndetection approaches, LFR can be applied to both batch and stream data; is not\nlimited by the distribution properties of the response variable (e.g., datasets\nwith imbalanced labels); is independent of the underlying statistical-model;\nand uses user-specified parameters that are intuitively comprehensible. The\nperformance of LFR is compared to benchmark approaches using both simulated and\ncommonly used public datasets that span the gamut of concept drift types. The\nresults show LFR significantly outperforms benchmark approaches in terms of\nrecall, accuracy and delay in detection of concept drifts across datasets.","url_abs":"http://arxiv.org/abs/1504.01044v2","url_pdf":"http://arxiv.org/pdf/1504.01044v2.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":"concept-drift-detection-for-streaming-data","repo_url":"https://github.com/mitre/menelaus/blob/dev/src/menelaus/concept_drift/lfr.py","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"drift-detection","task_name":"Drift Detection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1504.01044","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}