Analysing the classification of imbalanced data-sets with multiple classes: Binarization techniques and ad-hoc approaches

By Alberto Fernández, Victoria López, Mikel Galar, M. J. Jesús, Francisco Herrera. Knowledge-Based Systems, 2013

390 citations recorded by the source corpus.

Subfields: Accounting, Artificial Intelligence, Electrical and Electronic Engineering

View Paper: https://www.semanticscholar.org/paper/63e76c2c15031cd3686ac78330156d5e0aa6be1f

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

Learning from imbalanced data: open challenges and future directions (2016, methodology): “• Decomposition methods used so far [14] apply identical approach for each pair of classes.”
Classification with class imbalance problem: A review (2015, background): “This most likely happen because of two reasons; 1) class imbalanced dataset with binary class is more prevalent in most domain applications like in anomaly detection and 2) the higher degree of complication when multiple class imbalance problem is concerned [33][137].”
COVID-19 identification in chest X-ray images on flat and hierarchical classification scenarios (2020, neutral): “In order to deal with the imbalancess issue in classiﬁcation datasets, several methods have been proposed in the literature [22] and data level solutions are the most well-known and used techniques.”
To Combat Multi-Class Imbalanced Problems by Means of Over-Sampling Techniques (2016, methodology): “The advantages of sampling techniques are that they are more versatile and their use is independent of the classifiers selected [19].”
An insight into classification with imbalanced data: Empirical results and current trends on using data intrinsic characteristics (2013, background): “In most cases, the imbalanced class problem is associated to binary classification, but the multi-class problem often occurs and, since there can be several minority classes, it is more difficult to solve [48] [81].”
Boosting methods for multi-class imbalanced data classification: an experimental review (2020, background): “A multi-class problem can have different styles, including many minority and majority classes, one minority class and many minority classes or one majority class and many minority classes [24].”
MLSMOTE: Approaching imbalanced multilabel learning through synthetic instance generation (2015, background): “The other two approaches, algorithmic adaptations [37] and cost sensitive classification [38], are classifier dependent.”