The Class Imbalance Problem: Significance and Strategies

By Nathalie Japkowicz. 2000

792 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.

Citations in the corpus

Using Pre-Training Can Improve Model Robustness and Uncertainty (2019, background): “Cost sensitive learning (Japkowicz, 2000) balances the loss function by re-weighting each sample by the inverse frequency of its class.”
A novel ensemble method for classifying imbalanced data (2015, background): “Japkowicz [55] mainly discussed two strategies: under-sampling and resampling.”
A balanced accuracy function for epistasis modeling in imbalanced datasets using multifactor dimensionality reduction (2007, background): “Studies by Chawla et al. [2003, 2004] and Japkowicz [2000] provide extensive examples of these resampling techniques.”
Combating imbalance in network intrusion datasets (2006, background): “Over and under-sampling methodologies have received significant attention to counter the effect of imbalanced datasets [10], [17]–[20].”
Ensemble learning for credit card fraud detection (2018, background): “Japkowicz [11],[10], [5] discussed in detail the applicability of undersampling approach for handling the class-imbalance problem.”
SMOTEBoost: Improving Prediction of the Minority Class in Boosting (2003, background): “Several approaches for dealing with imbalanced data sets were recently introduced [1, 2, 4, 9-15].”
Data Augmentation for Deep-Learning-Based Electroencephalography. (2020, methodology): “There are various resampling strategies to solve this type of problem such as random under sampling, random oversampling and random under sampling with synthetic minority oversampling technique(SMOTE) [100-102].”
Learning from Imbalanced Data Sets: A Comparison of Various Strategies * (2000, background): “An additional study (reported in (Japkowicz 2000)) showed further that using more sophisticated over-sampling or down-sizing methods than a simple uniformly random approach appears unnecessary (at least, in the case of feedforward neural networks and simple artificial domains) but that the…”
Training Multi-Speaker Neural Text-to-Speech Systems using Speaker-Imbalanced Speech Corpora (2019, background): “we can choose to perform under-sampling [18] of the majority speakers, over-sampling of the minority speakers [19], or a little of both [15].”
Wrapper-based computation and evaluation of sampling methods for imbalanced datasets (2005, background): “[22] N. Japkowicz and S. Stephen.”
Applying Support Vector Machines to Imbalanced Datasets (2004, methodology): “shown, nevertheless, to make the classifier perform better than with the regular imbalanced sample [6].”
Data Mining for Imbalanced Datasets: An Overview (2005, methodology): “Over and under-sampling methodologies have received significant attention to counter the effect of imbalanced data sets (Solberg and Solberg, 1996; Japkowicz , 2000a; Chawla et al., 2002; Weiss and Provost, 2003; Kubat and Matwin, 1997; Jo and Japkowicz, 2004; Batista et al., 2004; Phua and…”
Towards Intersectionality in Machine Learning: Including More Identities, Handling Underrepresentation, and Performing Evaluation (2022, unknown): “A common machine learning technique for dealing with imbalanced classes is simply reweighting or resampling [68, 84].”
DORO: Distributional and Outlier Robust Optimization (2021, unknown): “…et al., 2012; Barocas & Selbst, 2016) where we want to test model’s performance on key demographic subpopulations, and learning with class imbalance (Japkowicz, 2000; Galar et al., 2011) where we train a clas-siﬁer on an imbalanced dataset with some minority classes having much fewer samples than…”
Revealing Household Characteristics from Smart Meter Data (2014, methodology): “An effective method to deal with class imbalance consists in undersampling the data during the training process [36, 37].”