Weighted online sequential extreme learning machine for class imbalance learning

Bilal Mirza, Zhiping Lin, Kar Ann Toh

Research output: Contribution to journalArticle

62 Citations (Scopus)

Abstract

Most of the existing sequential learning methods for class imbalance learn data in chunks. In this paper, we propose a weighted online sequential extreme learning machine (WOS-ELM) algorithm for class imbalance learning (CIL). WOS-ELM is a general online learning method that alleviates the class imbalance problem in both chunk-by-chunk and one-by-one learning. One of the new features of WOS-ELM is that an appropriate weight setting for CIL is selected in a computationally efficient manner. In one-by-one learning of WOS-ELM, a new sample can update the classification model without waiting for a chunk to be completed. Extensive empirical evaluations on 15 imbalanced datasets show that WOS-ELM obtains comparable or better classification performance than competing methods. The computational time of WOS-ELM is also found to be lower than that of the competing CIL methods.

Original languageEnglish
Pages (from-to)465-486
Number of pages22
JournalNeural Processing Letters
Volume38
Issue number3
DOIs
Publication statusPublished - 2013 Dec 1

Fingerprint

Learning systems
Learning
Machine Learning
Weights and Measures

All Science Journal Classification (ASJC) codes

  • Software
  • Neuroscience(all)
  • Computer Networks and Communications
  • Artificial Intelligence

Cite this

@article{62680a050a9444a98ece329a23e6d473,
title = "Weighted online sequential extreme learning machine for class imbalance learning",
abstract = "Most of the existing sequential learning methods for class imbalance learn data in chunks. In this paper, we propose a weighted online sequential extreme learning machine (WOS-ELM) algorithm for class imbalance learning (CIL). WOS-ELM is a general online learning method that alleviates the class imbalance problem in both chunk-by-chunk and one-by-one learning. One of the new features of WOS-ELM is that an appropriate weight setting for CIL is selected in a computationally efficient manner. In one-by-one learning of WOS-ELM, a new sample can update the classification model without waiting for a chunk to be completed. Extensive empirical evaluations on 15 imbalanced datasets show that WOS-ELM obtains comparable or better classification performance than competing methods. The computational time of WOS-ELM is also found to be lower than that of the competing CIL methods.",
author = "Bilal Mirza and Zhiping Lin and Toh, {Kar Ann}",
year = "2013",
month = "12",
day = "1",
doi = "10.1007/s11063-013-9286-9",
language = "English",
volume = "38",
pages = "465--486",
journal = "Neural Processing Letters",
issn = "1370-4621",
publisher = "Springer Netherlands",
number = "3",

}

Weighted online sequential extreme learning machine for class imbalance learning. / Mirza, Bilal; Lin, Zhiping; Toh, Kar Ann.

In: Neural Processing Letters, Vol. 38, No. 3, 01.12.2013, p. 465-486.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Weighted online sequential extreme learning machine for class imbalance learning

AU - Mirza, Bilal

AU - Lin, Zhiping

AU - Toh, Kar Ann

PY - 2013/12/1

Y1 - 2013/12/1

N2 - Most of the existing sequential learning methods for class imbalance learn data in chunks. In this paper, we propose a weighted online sequential extreme learning machine (WOS-ELM) algorithm for class imbalance learning (CIL). WOS-ELM is a general online learning method that alleviates the class imbalance problem in both chunk-by-chunk and one-by-one learning. One of the new features of WOS-ELM is that an appropriate weight setting for CIL is selected in a computationally efficient manner. In one-by-one learning of WOS-ELM, a new sample can update the classification model without waiting for a chunk to be completed. Extensive empirical evaluations on 15 imbalanced datasets show that WOS-ELM obtains comparable or better classification performance than competing methods. The computational time of WOS-ELM is also found to be lower than that of the competing CIL methods.

AB - Most of the existing sequential learning methods for class imbalance learn data in chunks. In this paper, we propose a weighted online sequential extreme learning machine (WOS-ELM) algorithm for class imbalance learning (CIL). WOS-ELM is a general online learning method that alleviates the class imbalance problem in both chunk-by-chunk and one-by-one learning. One of the new features of WOS-ELM is that an appropriate weight setting for CIL is selected in a computationally efficient manner. In one-by-one learning of WOS-ELM, a new sample can update the classification model without waiting for a chunk to be completed. Extensive empirical evaluations on 15 imbalanced datasets show that WOS-ELM obtains comparable or better classification performance than competing methods. The computational time of WOS-ELM is also found to be lower than that of the competing CIL methods.

UR - http://www.scopus.com/inward/record.url?scp=84890560408&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84890560408&partnerID=8YFLogxK

U2 - 10.1007/s11063-013-9286-9

DO - 10.1007/s11063-013-9286-9

M3 - Article

VL - 38

SP - 465

EP - 486

JO - Neural Processing Letters

JF - Neural Processing Letters

SN - 1370-4621

IS - 3

ER -