Inverse random under sampling for class imbalance problem and its application to multi-label classification

Tahir, MA, Kittler, J and Yan, F (2012) Inverse random under sampling for class imbalance problem and its application to multi-label classification Pattern Recognition, 45 (10). pp. 3738-3750.

	PDF PR12.pdf Restricted to Repository staff only Available under License : See the attached licence file. Download (1MB)
	PDF (licence) SRI_deposit_agreement.pdf Restricted to Repository staff only Download (33kB)

Abstract

In this paper, a novel inverse random undersampling (IRUS) method is proposed for the class imbalance problem. The main idea is to severely under sample the majority class thus creating a large number of distinct training sets. For each training set we then find a decision boundary which separates the minority class from the majority class. By combining the multiple designs through fusion, we construct a composite boundary between the majority class and the minority class. The proposed methodology is applied on 22 UCI data sets and experimental results indicate a significant increase in performance when compared with many existing class-imbalance learning methods. We also present promising results for multi-label classification, a challenging research problem in many modern applications such as music, text and image categorization.

Item Type:	Article
Authors :	Authors Email ORCID Tahir, MA UNSPECIFIED UNSPECIFIED Kittler, J UNSPECIFIED UNSPECIFIED Yan, F UNSPECIFIED UNSPECIFIED
Date :	2012
Identification Number :	https://doi.org/10.1016/j.patcog.2012.03.014
Depositing User :	Symplectic Elements
Date Deposited :	28 Mar 2017 14:11
Last Modified :	28 Mar 2017 14:11
URI:	http://epubs.surrey.ac.uk/id/eprint/733256

Actions (login required)

View Item

Downloads