EPRENNID: An evolutionary prototype reduction based ensemble for nearest neighbor classification of imbalanced data

Vluymans, Sarah and Triguero, Isaac and Cornelis, Chris and Saeys, Yvan (2016) EPRENNID: An evolutionary prototype reduction based ensemble for nearest neighbor classification of imbalanced data. Neurocomputing, 216 . pp. 596-610. ISSN 0925-2312

[img] PDF - Repository staff only until 11 August 2017. - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Available under Licence Creative Commons Attribution Non-commercial No Derivatives.
Download (925kB)

Abstract

Classification problems with an imbalanced class distribution have received an increased amount of attention within the machine learning community over the last decade. They are encountered in a growing number of real-world situations and pose a challenge to standard machine learning techniques. We propose a new hybrid method specifically tailored to handle class imbalance, called EPRENNID. It performs an evolutionary prototype reduction focused on providing diverse solutions to prevent the method from overfitting the training set. It also allows us to explicitly reduce the underrepresented class, which the most common preprocessing solutions handling class imbalance usually protect. As part of the experimental study, we show that the proposed prototype reduction method outperforms state-of-the-art preprocessing techniques. The preprocessing step yields multiple prototype sets that are later used in an ensemble, performing a weighted voting scheme with the nearest neighbor classifier. EPRENNID is experimentally shown to significantly outperform previous proposals.

Item Type: Article
Keywords: Imbalanced data; Prototype selection; Prototype generation; Differential evolution; Nearest neighbor
Schools/Departments: University of Nottingham, UK > Faculty of Science > School of Computer Science
Identification Number: https://doi.org/10.1016/j.neucom.2016.08.026
Depositing User: Blythe, Mrs Maxine
Date Deposited: 26 Aug 2016 12:07
Last Modified: 16 Nov 2016 19:29
URI: http://eprints.nottingham.ac.uk/id/eprint/36055

Actions (Archive Staff Only)

Edit View Edit View