On the effectiveness of preprocessing methods when dealing with different levels of class imbalance

By Vicente García, José Salvador Sánchez, Ramón Alberto Mollineda. Knowledge-Based Systems, 2012

359 citations recorded by the source corpus.

Subfields: Accounting, Artificial Intelligence, Electrical and Electronic Engineering

View Paper: https://www.semanticscholar.org/paper/76024a2772ab7ebbccc472e7b44539860a0091da

How researchers see this paper

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

In Accounting

In Artificial Intelligence

In Electrical and Electronic Engineering

Citations in the corpus

MLSMOTE: Approaching imbalanced multilabel learning through synthetic instance generation (2015, background): “Resampling techniques are classifier-independent solutions to the imbalanced learning problem, albeit some proposals for specific classifiers exist [36], and have shown their general effectiveness [14].”
To Combat Multi-Class Imbalanced Problems by Means of Over-Sampling Techniques (2016, background): “[24] indicates that “over-sampling the minority class consistently outperforms under-sampling the majority class when data sets are strongly imbalanced, whereas there are not significant differences for databases with a low imbalance”.”
Dual autoencoders features for imbalance classification problem (2016, background): “, data distribution) while oversampling-based methods may suffer from overfitting and becomes sensitive to overlapping between minority and majority classes [1].”
An insight into classification with imbalanced data: Empirical results and current trends on using data intrinsic characteristics (2013, background): “Most of the studies on the behavior of several standard classifiers in imbalance domains have shown that significant loss of performance is mainly due to the skewed class distribution, given by the imbalance ratio (IR), defined as the ratio of the number of instances in the majority class to the num”
Analysing the classification of imbalanced data-sets with multiple classes: Binarization techniques and ad-hoc approaches (2013, methodology): “In this work we use the imbalance ratio (IR) [44], deﬁned as the ratio of the number of instances of the majority (known as the negative class) and the minority class (known as the positive class), to organise the different data-sets according to this measure [30].”