Customizing asynchronous parallel pattern search algorithm to improve ANN classifier for learning disabilities students identification

T. K. Wu, S. C. Huang, W. W. Chiou, Y. R. Meng

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

Due to the implicit characteristics of learning disabilities (LD), the diagnosis of students with learning disabilities has been a difficult process that requires extensive man power and takes a long time. Through genetic-based parameters optimization, artificial neural network (ANN) classifier has proven to be a good predictor to the diagnosis of students with learning disabilities. In this study, we examine another optimization algorithm, the asynchronous parallel pattern search (APPS), to search for appropriate parameters in constructing ANN-based LD classifier. To fully take advantage of modern multi-cored CPU technologies and to further expand the potential search space, various modifications to both of the serial and parallel versions of the original APPS implementations have been developed. The outcomes show that APPS in its original implementation can be competitive to genetic algorithm in term of accuracy, while requiring much less execution time. Furthermore, with consecutive (two-step) applications of the modified APPS algorithm to fine-tune the ANN parameters, we have further improved the ANN-based LD identification accuracy as compared to our previous results using genetic algorithm.

Original languageEnglish
Title of host publicationProceedings - 2011 7th International Conference on Natural Computation, ICNC 2011
Pages1639-1643
Number of pages5
DOIs
Publication statusPublished - 2011 Oct 6
Event2011 7th International Conference on Natural Computation, ICNC 2011 - Shanghai, China
Duration: 2011 Jul 262011 Jul 28

Publication series

NameProceedings - 2011 7th International Conference on Natural Computation, ICNC 2011
Volume3

Other

Other2011 7th International Conference on Natural Computation, ICNC 2011
CountryChina
CityShanghai
Period11-07-2611-07-28

Fingerprint

Learning Disorders
Classifiers
Students
Neural networks
Genetic algorithms
Program processors
Identification (Psychology)
Technology

All Science Journal Classification (ASJC) codes

  • Computational Theory and Mathematics
  • Neuroscience(all)

Cite this

Wu, T. K., Huang, S. C., Chiou, W. W., & Meng, Y. R. (2011). Customizing asynchronous parallel pattern search algorithm to improve ANN classifier for learning disabilities students identification. In Proceedings - 2011 7th International Conference on Natural Computation, ICNC 2011 (pp. 1639-1643). [6022322] (Proceedings - 2011 7th International Conference on Natural Computation, ICNC 2011; Vol. 3). https://doi.org/10.1109/ICNC.2011.6022322
Wu, T. K. ; Huang, S. C. ; Chiou, W. W. ; Meng, Y. R. / Customizing asynchronous parallel pattern search algorithm to improve ANN classifier for learning disabilities students identification. Proceedings - 2011 7th International Conference on Natural Computation, ICNC 2011. 2011. pp. 1639-1643 (Proceedings - 2011 7th International Conference on Natural Computation, ICNC 2011).
@inproceedings{c306ffc0b61948d5956d9e9b56385687,
title = "Customizing asynchronous parallel pattern search algorithm to improve ANN classifier for learning disabilities students identification",
abstract = "Due to the implicit characteristics of learning disabilities (LD), the diagnosis of students with learning disabilities has been a difficult process that requires extensive man power and takes a long time. Through genetic-based parameters optimization, artificial neural network (ANN) classifier has proven to be a good predictor to the diagnosis of students with learning disabilities. In this study, we examine another optimization algorithm, the asynchronous parallel pattern search (APPS), to search for appropriate parameters in constructing ANN-based LD classifier. To fully take advantage of modern multi-cored CPU technologies and to further expand the potential search space, various modifications to both of the serial and parallel versions of the original APPS implementations have been developed. The outcomes show that APPS in its original implementation can be competitive to genetic algorithm in term of accuracy, while requiring much less execution time. Furthermore, with consecutive (two-step) applications of the modified APPS algorithm to fine-tune the ANN parameters, we have further improved the ANN-based LD identification accuracy as compared to our previous results using genetic algorithm.",
author = "Wu, {T. K.} and Huang, {S. C.} and Chiou, {W. W.} and Meng, {Y. R.}",
year = "2011",
month = "10",
day = "6",
doi = "10.1109/ICNC.2011.6022322",
language = "English",
isbn = "9781424499533",
series = "Proceedings - 2011 7th International Conference on Natural Computation, ICNC 2011",
pages = "1639--1643",
booktitle = "Proceedings - 2011 7th International Conference on Natural Computation, ICNC 2011",

}

Wu, TK, Huang, SC, Chiou, WW & Meng, YR 2011, Customizing asynchronous parallel pattern search algorithm to improve ANN classifier for learning disabilities students identification. in Proceedings - 2011 7th International Conference on Natural Computation, ICNC 2011., 6022322, Proceedings - 2011 7th International Conference on Natural Computation, ICNC 2011, vol. 3, pp. 1639-1643, 2011 7th International Conference on Natural Computation, ICNC 2011, Shanghai, China, 11-07-26. https://doi.org/10.1109/ICNC.2011.6022322

Customizing asynchronous parallel pattern search algorithm to improve ANN classifier for learning disabilities students identification. / Wu, T. K.; Huang, S. C.; Chiou, W. W.; Meng, Y. R.

Proceedings - 2011 7th International Conference on Natural Computation, ICNC 2011. 2011. p. 1639-1643 6022322 (Proceedings - 2011 7th International Conference on Natural Computation, ICNC 2011; Vol. 3).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Customizing asynchronous parallel pattern search algorithm to improve ANN classifier for learning disabilities students identification

AU - Wu, T. K.

AU - Huang, S. C.

AU - Chiou, W. W.

AU - Meng, Y. R.

PY - 2011/10/6

Y1 - 2011/10/6

N2 - Due to the implicit characteristics of learning disabilities (LD), the diagnosis of students with learning disabilities has been a difficult process that requires extensive man power and takes a long time. Through genetic-based parameters optimization, artificial neural network (ANN) classifier has proven to be a good predictor to the diagnosis of students with learning disabilities. In this study, we examine another optimization algorithm, the asynchronous parallel pattern search (APPS), to search for appropriate parameters in constructing ANN-based LD classifier. To fully take advantage of modern multi-cored CPU technologies and to further expand the potential search space, various modifications to both of the serial and parallel versions of the original APPS implementations have been developed. The outcomes show that APPS in its original implementation can be competitive to genetic algorithm in term of accuracy, while requiring much less execution time. Furthermore, with consecutive (two-step) applications of the modified APPS algorithm to fine-tune the ANN parameters, we have further improved the ANN-based LD identification accuracy as compared to our previous results using genetic algorithm.

AB - Due to the implicit characteristics of learning disabilities (LD), the diagnosis of students with learning disabilities has been a difficult process that requires extensive man power and takes a long time. Through genetic-based parameters optimization, artificial neural network (ANN) classifier has proven to be a good predictor to the diagnosis of students with learning disabilities. In this study, we examine another optimization algorithm, the asynchronous parallel pattern search (APPS), to search for appropriate parameters in constructing ANN-based LD classifier. To fully take advantage of modern multi-cored CPU technologies and to further expand the potential search space, various modifications to both of the serial and parallel versions of the original APPS implementations have been developed. The outcomes show that APPS in its original implementation can be competitive to genetic algorithm in term of accuracy, while requiring much less execution time. Furthermore, with consecutive (two-step) applications of the modified APPS algorithm to fine-tune the ANN parameters, we have further improved the ANN-based LD identification accuracy as compared to our previous results using genetic algorithm.

UR - http://www.scopus.com/inward/record.url?scp=80053412981&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80053412981&partnerID=8YFLogxK

U2 - 10.1109/ICNC.2011.6022322

DO - 10.1109/ICNC.2011.6022322

M3 - Conference contribution

AN - SCOPUS:80053412981

SN - 9781424499533

T3 - Proceedings - 2011 7th International Conference on Natural Computation, ICNC 2011

SP - 1639

EP - 1643

BT - Proceedings - 2011 7th International Conference on Natural Computation, ICNC 2011

ER -

Wu TK, Huang SC, Chiou WW, Meng YR. Customizing asynchronous parallel pattern search algorithm to improve ANN classifier for learning disabilities students identification. In Proceedings - 2011 7th International Conference on Natural Computation, ICNC 2011. 2011. p. 1639-1643. 6022322. (Proceedings - 2011 7th International Conference on Natural Computation, ICNC 2011). https://doi.org/10.1109/ICNC.2011.6022322