An Optimal Approach for Pruning Annular Regularized Extreme Learning Machines

Lavneet Singh, Girija CHETTY

Research output: A Conference proceeding or a Chapter in BookConference contribution

1 Citation (Scopus)

Abstract

Larger datasets, with many samples are problematic for solving problems in data mining and machine learning, due to increase in computational times, increased complexity, and bad generalization due to outliers. Further, the accuracy and performance of machine learning and statistical models are still based on tuning of some parameters and optimizing them for generating better predictive models of learning. In this paper, we propose a novel formulation of Extreme Learning Machines - the Annular ELM, with RANSAC multi model response regularization for pruning large number of hidden nodes to acquire better optimality, generalization and classification accuracy. Experimental evaluation of the proposed ELM formulation on different benchmark datasets showed that the algorithm optimally prunes the hidden nodes, with better generalization and higher classification accuracy as compared to other algorithms, including the well-known SVM, OP-ELM for binary and multi-class classification and regression problems. Also, we extended the proposed algorithm to a more complex application context involving MRI Brain Image classification. For this study, we examine the performance of the proposed algorithm on magnetic resonance images (MRI) of various states of brain by extracting the most significant features, and to classify them into normal and abnormal brain images.
Original languageEnglish
Title of host publication2014 IEEE International Conference on Data Mining Workshop (ICDMW)
EditorsZhi-Hua Zhou, Wei Wang, Ravi Kumar, Hannu Toivonen, Jian Pei, Joshua Zhexue Huang, Xindong Wu
Place of PublicationShenzhen, China
PublisherIEEE
Pages80-87
Number of pages8
ISBN (Electronic)9781479942749
ISBN (Print)9781479942756
DOIs
Publication statusPublished - 2014
Event2014 IEEE International Conference on Data Mining - Shenzen, Shenzen, China
Duration: 14 Dec 2014 → …

Publication series

NameIEEE International Conference on Data Mining Workshops, ICDMW
NumberJanuary
Volume2015-January
ISSN (Print)2375-9232
ISSN (Electronic)2375-9259

Conference

Conference2014 IEEE International Conference on Data Mining
CountryChina
CityShenzen
Period14/12/14 → …

Fingerprint

Learning systems
Brain
Magnetic resonance
Image classification
Data mining
Tuning

Cite this

Singh, L., & CHETTY, G. (2014). An Optimal Approach for Pruning Annular Regularized Extreme Learning Machines. In Z-H. Zhou, W. Wang, R. Kumar, H. Toivonen, J. Pei, J. Zhexue Huang, & X. Wu (Eds.), 2014 IEEE International Conference on Data Mining Workshop (ICDMW) (pp. 80-87). [7022582] (IEEE International Conference on Data Mining Workshops, ICDMW; Vol. 2015-January, No. January). Shenzhen, China: IEEE. https://doi.org/10.1109/ICDMW.2014.69
Singh, Lavneet ; CHETTY, Girija. / An Optimal Approach for Pruning Annular Regularized Extreme Learning Machines. 2014 IEEE International Conference on Data Mining Workshop (ICDMW). editor / Zhi-Hua Zhou ; Wei Wang ; Ravi Kumar ; Hannu Toivonen ; Jian Pei ; Joshua Zhexue Huang ; Xindong Wu. Shenzhen, China : IEEE, 2014. pp. 80-87 (IEEE International Conference on Data Mining Workshops, ICDMW; January).
@inproceedings{9f0f1a95f8e64c90b72b22ea96bf45cb,
title = "An Optimal Approach for Pruning Annular Regularized Extreme Learning Machines",
abstract = "Larger datasets, with many samples are problematic for solving problems in data mining and machine learning, due to increase in computational times, increased complexity, and bad generalization due to outliers. Further, the accuracy and performance of machine learning and statistical models are still based on tuning of some parameters and optimizing them for generating better predictive models of learning. In this paper, we propose a novel formulation of Extreme Learning Machines - the Annular ELM, with RANSAC multi model response regularization for pruning large number of hidden nodes to acquire better optimality, generalization and classification accuracy. Experimental evaluation of the proposed ELM formulation on different benchmark datasets showed that the algorithm optimally prunes the hidden nodes, with better generalization and higher classification accuracy as compared to other algorithms, including the well-known SVM, OP-ELM for binary and multi-class classification and regression problems. Also, we extended the proposed algorithm to a more complex application context involving MRI Brain Image classification. For this study, we examine the performance of the proposed algorithm on magnetic resonance images (MRI) of various states of brain by extracting the most significant features, and to classify them into normal and abnormal brain images.",
keywords = "Classification, Extreme Learning Machine, MRI Images, RANSAC, Regularization",
author = "Lavneet Singh and Girija CHETTY",
year = "2014",
doi = "10.1109/ICDMW.2014.69",
language = "English",
isbn = "9781479942756",
series = "IEEE International Conference on Data Mining Workshops, ICDMW",
publisher = "IEEE",
number = "January",
pages = "80--87",
editor = "Zhi-Hua Zhou and Wei Wang and Ravi Kumar and Hannu Toivonen and Jian Pei and {Zhexue Huang}, Joshua and Xindong Wu",
booktitle = "2014 IEEE International Conference on Data Mining Workshop (ICDMW)",

}

Singh, L & CHETTY, G 2014, An Optimal Approach for Pruning Annular Regularized Extreme Learning Machines. in Z-H Zhou, W Wang, R Kumar, H Toivonen, J Pei, J Zhexue Huang & X Wu (eds), 2014 IEEE International Conference on Data Mining Workshop (ICDMW)., 7022582, IEEE International Conference on Data Mining Workshops, ICDMW, no. January, vol. 2015-January, IEEE, Shenzhen, China, pp. 80-87, 2014 IEEE International Conference on Data Mining, Shenzen, China, 14/12/14. https://doi.org/10.1109/ICDMW.2014.69

An Optimal Approach for Pruning Annular Regularized Extreme Learning Machines. / Singh, Lavneet; CHETTY, Girija.

2014 IEEE International Conference on Data Mining Workshop (ICDMW). ed. / Zhi-Hua Zhou; Wei Wang; Ravi Kumar; Hannu Toivonen; Jian Pei; Joshua Zhexue Huang; Xindong Wu. Shenzhen, China : IEEE, 2014. p. 80-87 7022582 (IEEE International Conference on Data Mining Workshops, ICDMW; Vol. 2015-January, No. January).

Research output: A Conference proceeding or a Chapter in BookConference contribution

TY - GEN

T1 - An Optimal Approach for Pruning Annular Regularized Extreme Learning Machines

AU - Singh, Lavneet

AU - CHETTY, Girija

PY - 2014

Y1 - 2014

N2 - Larger datasets, with many samples are problematic for solving problems in data mining and machine learning, due to increase in computational times, increased complexity, and bad generalization due to outliers. Further, the accuracy and performance of machine learning and statistical models are still based on tuning of some parameters and optimizing them for generating better predictive models of learning. In this paper, we propose a novel formulation of Extreme Learning Machines - the Annular ELM, with RANSAC multi model response regularization for pruning large number of hidden nodes to acquire better optimality, generalization and classification accuracy. Experimental evaluation of the proposed ELM formulation on different benchmark datasets showed that the algorithm optimally prunes the hidden nodes, with better generalization and higher classification accuracy as compared to other algorithms, including the well-known SVM, OP-ELM for binary and multi-class classification and regression problems. Also, we extended the proposed algorithm to a more complex application context involving MRI Brain Image classification. For this study, we examine the performance of the proposed algorithm on magnetic resonance images (MRI) of various states of brain by extracting the most significant features, and to classify them into normal and abnormal brain images.

AB - Larger datasets, with many samples are problematic for solving problems in data mining and machine learning, due to increase in computational times, increased complexity, and bad generalization due to outliers. Further, the accuracy and performance of machine learning and statistical models are still based on tuning of some parameters and optimizing them for generating better predictive models of learning. In this paper, we propose a novel formulation of Extreme Learning Machines - the Annular ELM, with RANSAC multi model response regularization for pruning large number of hidden nodes to acquire better optimality, generalization and classification accuracy. Experimental evaluation of the proposed ELM formulation on different benchmark datasets showed that the algorithm optimally prunes the hidden nodes, with better generalization and higher classification accuracy as compared to other algorithms, including the well-known SVM, OP-ELM for binary and multi-class classification and regression problems. Also, we extended the proposed algorithm to a more complex application context involving MRI Brain Image classification. For this study, we examine the performance of the proposed algorithm on magnetic resonance images (MRI) of various states of brain by extracting the most significant features, and to classify them into normal and abnormal brain images.

KW - Classification

KW - Extreme Learning Machine

KW - MRI Images

KW - RANSAC

KW - Regularization

UR - http://www.scopus.com/inward/record.url?scp=84936873249&partnerID=8YFLogxK

U2 - 10.1109/ICDMW.2014.69

DO - 10.1109/ICDMW.2014.69

M3 - Conference contribution

SN - 9781479942756

T3 - IEEE International Conference on Data Mining Workshops, ICDMW

SP - 80

EP - 87

BT - 2014 IEEE International Conference on Data Mining Workshop (ICDMW)

A2 - Zhou, Zhi-Hua

A2 - Wang, Wei

A2 - Kumar, Ravi

A2 - Toivonen, Hannu

A2 - Pei, Jian

A2 - Zhexue Huang, Joshua

A2 - Wu, Xindong

PB - IEEE

CY - Shenzhen, China

ER -

Singh L, CHETTY G. An Optimal Approach for Pruning Annular Regularized Extreme Learning Machines. In Zhou Z-H, Wang W, Kumar R, Toivonen H, Pei J, Zhexue Huang J, Wu X, editors, 2014 IEEE International Conference on Data Mining Workshop (ICDMW). Shenzhen, China: IEEE. 2014. p. 80-87. 7022582. (IEEE International Conference on Data Mining Workshops, ICDMW; January). https://doi.org/10.1109/ICDMW.2014.69