Learning in the presence of concept drift and hidden contexts

By Gerhard Widmer, Miroslav Kubat. 2004

407 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

Citations in the corpus

Domain Generalization via Invariant Feature Representation (2013, background): “In this case, the data exhibits covariate (or dataset) shift (Widmer and Kurat 1996, Quionero-Candela et al. 2009, Bickel et al. 2009b): although the marginal distributions PX on cell attributes vary due to biological or technical variations, the functional relationship P(Y”
A review of ensemble learning and data augmentation models for class imbalanced problems: combination, implementation and evaluation (2023, background): “Concept drift is another issue that arises for online learners, where changes in the underlying distributions of the incoming data [292, 293] require models to quickly track such changes and adapt to them.”
From Concept Drift to Model Degradation: An Overview on Performance-Aware Drift Detectors (2022, background): “The first type is caused by changes of unknown context that cannot be measured or represented in the available attributes of the dataset, which is known as hidden context [3].”
Classifier Ensembles for Detecting Concept Change in Streaming Data: Overview and Perspectives (2008, background): “Many single classifier models also eva luate the accuracy either to select the window size for the next classifier [2,10,18–20,23] or to update the current model [1,15,41].”
Incremental Clustering for the Classification of Concept-Drifting Data Streams (2008, background): “Examples of this group can be found in [5, 10, 19].”
Adaptive Machine Learning for Credit Card Fraud Detection (2015, methodology): “FLORA approach [164] uses a variable forgetting rate where the window is shrunk if a change is detected and expanded otherwise.”
An Online Learning Approach to Interpolation and Extrapolation in Domain Generalization (2021, background): “One common choice is to make specific assumptions about covariate or label shift (Widmer & Kubat, 1996; Bickel et al., 2009; Lipton et al., 2018); another approach is Distributionally Robust Optimization (DRO), where the test distribution is assumed to lie in some uncertainty set around the…”
Machine learning and deep learning (2021, methodology): “While techniques for automated learning exist that involve using trusted data windows and concept descriptions (Widmer and Kubat 1996), automated strategies for discovering and solving business-related problems are a challenge (Pentland et al.”
A survey on fake news and rumour detection techniques (2019, background): “Second, given the streaming nature of social media, information could be subject to concept drift [102].”