Skip to main content

Research Repository

See what's under the surface

Neighbourhood-based undersampling approach for handling imbalanced and overlapped data.

Vuttipittayamongkol, Pattaramon; Elyan, Eyad

Authors

Pattaramon Vuttipittayamongkol

Eyad Elyan



Abstract

Class imbalanced datasets are common across different domains including health, security, banking and others. A typical supervised learning algorithm tends to be biased towards the majority class when dealing with imbalanced datasets. The learning task becomes more challenging when there is also an overlap of instances from different classes. In this paper, we propose an undersampling framework for handling class imbalance in binary datasets by removing potential overlapped data points. Our methods are designed to identify and eliminate majority class instances from the overlapping region. Accurate identification and elimination of these instances maximise the visibility of the minority class instances and at the same time minimises excessive elimination of data, which reduces information loss. Four methods based on neighbourhood searching with different criteria to identify potential overlapped instances are proposed in this paper. Extensive experiments using simulated and real-world datasets were carried out. Results show comparable performance with state-of-the-art methods across different common metrics with exceptional and statistically significant improvements in sensitivity.

Journal Article Type Article
Publication Date Jan 31, 2020
Journal Information Sciences
Print ISSN 0020-0255
Publisher Elsevier
Peer Reviewed Peer Reviewed
Volume 509
Pages 47-70
Institution Citation VUTTIPITTAYAMONGKOL, P. and ELYAN, E. 2020. Neighbourhood-based undersampling approach for handling imbalanced and overlapped data. Information sciences [online], 509, pages 47-70. Available from: https://doi.org/10.1016/j.ins.2019.08.062
DOI https://doi.org/10.1016/j.ins.2019.08.062
Keywords Imbalanced dataset; Undersampling; k-NN; Class overlap; Classification

This file is under embargo due to copyright reasons.

Contact publications@rgu.ac.uk to request a copy for personal use.




Downloadable Citations