A Multiple Resampling Method for Learning from Imbalanced Data Sets

By Andrew Estabrooks, Taeho C. Jo, Nathalie Japkowicz. International Conference on Climate Informatics, 2004

1,076 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Electrical and Electronic Engineering

View Paper: https://www.semanticscholar.org/paper/1152d9bfb6cbfda1b919ff6e9013f48344f9926f

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

In Electrical and Electronic Engineering

Citations in the corpus

Analysing the classification of imbalanced data-sets with multiple classes: Binarization techniques and ad-hoc approaches (2013, background): “For example, solutions at data level [5,15,35] suffer from the increased search space, and solutions at algorithm level [66,31] become more complicated, as the learning algorithm must consider several small classes.”
A Closer Look At Feature Space Data Augmentation For Few-Shot Intent Classification (2019, methodology): “Upsampling is a well studied technique to handle the class imbalance problem (Estabrooks et al., 2004).”
An insight into classification with imbalanced data: Empirical results and current trends on using data intrinsic characteristics (2013, background): “These approaches can be categorized into two groups: the internal approaches that create new algorithms or modify existing ones to take the class-imbalance problem into consideration [7,41,82,129,152] and external approaches that preprocess the data in order to diminish the effect of their class imb”
A Survey of Predictive Modeling on Imbalanced Domains (2016, methodology): “One of the ﬁrst hybrid strategies was presented by Estabrooks and Japkowicz [2001] and Estabrooks et al. [2004].”
An introduction to domain adaptation and transfer learning (2018, background): “To avoid this unwanted behaviour, samples from the rare class could be re-weighted to be more important [62,37,63].”
On the Class Imbalance Problem (2008, methodology): “More detailed experiments are presented in [ 46 ].”
AUC-Maximized Deep Convolutional Neural Fields for Protein Sequence Labeling (2016, methodology): “These methods include (a) under-sampling the majority class; (b) over-sampling the minority class; or (c) combining both under-sampling and over-sampling [4,13,32].”
Calibrating Probability with Undersampling for Unbalanced Classification (2015, background): “Even if undersampling produces biased probability estimates, it is often used to balance datasets with skewed class distributions because several classifiers have empirically shown better performance when trained on balanced dataset [s5], [14].”
Adaptive Machine Learning for Credit Card Fraud Detection (2015, methodology): “In practice, sampling methods are often used to balance datasets with skewed class distributions because several classifiers have empirically shown better performance when trained on balanced dataset [88, 90].”
Oversampling for Imbalanced Learning Based on K-Means and SMOTE (2017, background): “The optimal imbalance ratio is not obvious and has been discussed by other researchers (Provost, 2000; Estabrooks et al., 2004).”
Gradient boosting trees for auto insurance loss cost modeling and prediction (2012, background): “In econometrics, this sample scheme is known as choice-based or endogenous stratified sampling (Green, 2000) and it is also popular in the computer science community (Chan & Stolfo, 1998; Estabrooks & Japkowicz, 2004).”
Learning from Imbalanced Data (2009, background): “Studies have shown that for several base classifiers, a balanced data set provides improved overall classification performance compared to an imbalanced data set [35], [36], [37].”
Analysis of preprocessing vs. cost-sensitive learning for imbalanced classification. Open problems on intrinsic data characteristics (2012, methodology): “To compute the AUC we just need to obtain the area under the curve as: AUC ¼ 1þ TPrate FPrate 2 ð2Þ”
Study of the impact of resampling methods for contrast pattern based classifiers in imbalanced databases (2016, background): “Although there are many studies about resampling methods, in these studies the authors do not determine which resampling method is the best, due to the fact that there is not a resampling method that always gets clearly the best results [62].”
On the effectiveness of preprocessing methods when dealing with different levels of class imbalance (2012, background): “Examples of the former are approaches to internally biasing the discriminating process [14] and multi-experts systems [21], whereas the data level solutions consist of artificially resampling the original data set until the problem classes are approximately equally represented.”
A study of the behaviour of linguistic fuzzy rule based classification systems in the framework of imbalanced data-sets (2008, background): “In the specialized literature, we can find some papers for re-sampling techniques from the study point of view of the effect of the class distribution in classification [43,13] and adaptations of prototype selection methods [46] to deal with imbalanced data-sets.”