Multivariate filter and PSO in protein function classification

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

Abstract

Protein features are often complex, and they are challenging to classify. In identifying the most discriminatory features in protein sequences, we propose a new feature-selection strategy by integrating the multivariate filter and Particle Swarm Optimisation (PSO) algorithms. Experimental results, based on the number of reducts and classification accuracy, were analysed in both the filter and wrapper phases. For our dataset, the proposed method statistically significantly improves the obtained classification accuracy and reduces the number of feature subsets. In the filter phase, the accuracy is improved more than 4% in three out of four multivariate feature selection methods compared to a model without feature selection. In the second phase, the accuracy is increased from 97.51% to 100%. We also demonstrate the importance of the correct parameter settings in the PSO to guarantee good performance.

Original languageEnglish
Title of host publicationProceedings of the 2010 International Conference of Soft Computing and Pattern Recognition, SoCPaR 2010
Pages330-333
Number of pages4
DOIs
Publication statusPublished - 2010
Event2010 International Conference of Soft Computing and Pattern Recognition, SoCPaR 2010 - Cergy-Pontoise
Duration: 7 Dec 201010 Dec 2010

Other

Other2010 International Conference of Soft Computing and Pattern Recognition, SoCPaR 2010
CityCergy-Pontoise
Period7/12/1010/12/10

Fingerprint

Particle swarm optimization (PSO)
Feature extraction
Proteins

Keywords

  • Feature selection
  • Multivariate filter
  • Particle swarm optimisation
  • Protein sequences

ASJC Scopus subject areas

  • Computational Theory and Mathematics
  • Computer Science Applications
  • Computer Vision and Pattern Recognition

Cite this

Abdul-Rahman, S., Mohamed Hussein, Z. A., & Abu Bakar, A. (2010). Multivariate filter and PSO in protein function classification. In Proceedings of the 2010 International Conference of Soft Computing and Pattern Recognition, SoCPaR 2010 (pp. 330-333). [5686158] https://doi.org/10.1109/SOCPAR.2010.5686158

Multivariate filter and PSO in protein function classification. / Abdul-Rahman, Shuzlina; Mohamed Hussein, Zeti Azura; Abu Bakar, Azuraliza.

Proceedings of the 2010 International Conference of Soft Computing and Pattern Recognition, SoCPaR 2010. 2010. p. 330-333 5686158.

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

Abdul-Rahman, S, Mohamed Hussein, ZA & Abu Bakar, A 2010, Multivariate filter and PSO in protein function classification. in Proceedings of the 2010 International Conference of Soft Computing and Pattern Recognition, SoCPaR 2010., 5686158, pp. 330-333, 2010 International Conference of Soft Computing and Pattern Recognition, SoCPaR 2010, Cergy-Pontoise, 7/12/10. https://doi.org/10.1109/SOCPAR.2010.5686158
Abdul-Rahman S, Mohamed Hussein ZA, Abu Bakar A. Multivariate filter and PSO in protein function classification. In Proceedings of the 2010 International Conference of Soft Computing and Pattern Recognition, SoCPaR 2010. 2010. p. 330-333. 5686158 https://doi.org/10.1109/SOCPAR.2010.5686158
Abdul-Rahman, Shuzlina ; Mohamed Hussein, Zeti Azura ; Abu Bakar, Azuraliza. / Multivariate filter and PSO in protein function classification. Proceedings of the 2010 International Conference of Soft Computing and Pattern Recognition, SoCPaR 2010. 2010. pp. 330-333
@inproceedings{2440859be7a84ee2ba39b4e73ebedee8,
title = "Multivariate filter and PSO in protein function classification",
abstract = "Protein features are often complex, and they are challenging to classify. In identifying the most discriminatory features in protein sequences, we propose a new feature-selection strategy by integrating the multivariate filter and Particle Swarm Optimisation (PSO) algorithms. Experimental results, based on the number of reducts and classification accuracy, were analysed in both the filter and wrapper phases. For our dataset, the proposed method statistically significantly improves the obtained classification accuracy and reduces the number of feature subsets. In the filter phase, the accuracy is improved more than 4{\%} in three out of four multivariate feature selection methods compared to a model without feature selection. In the second phase, the accuracy is increased from 97.51{\%} to 100{\%}. We also demonstrate the importance of the correct parameter settings in the PSO to guarantee good performance.",
keywords = "Feature selection, Multivariate filter, Particle swarm optimisation, Protein sequences",
author = "Shuzlina Abdul-Rahman and {Mohamed Hussein}, {Zeti Azura} and {Abu Bakar}, Azuraliza",
year = "2010",
doi = "10.1109/SOCPAR.2010.5686158",
language = "English",
isbn = "9781424478958",
pages = "330--333",
booktitle = "Proceedings of the 2010 International Conference of Soft Computing and Pattern Recognition, SoCPaR 2010",

}

TY - GEN

T1 - Multivariate filter and PSO in protein function classification

AU - Abdul-Rahman, Shuzlina

AU - Mohamed Hussein, Zeti Azura

AU - Abu Bakar, Azuraliza

PY - 2010

Y1 - 2010

N2 - Protein features are often complex, and they are challenging to classify. In identifying the most discriminatory features in protein sequences, we propose a new feature-selection strategy by integrating the multivariate filter and Particle Swarm Optimisation (PSO) algorithms. Experimental results, based on the number of reducts and classification accuracy, were analysed in both the filter and wrapper phases. For our dataset, the proposed method statistically significantly improves the obtained classification accuracy and reduces the number of feature subsets. In the filter phase, the accuracy is improved more than 4% in three out of four multivariate feature selection methods compared to a model without feature selection. In the second phase, the accuracy is increased from 97.51% to 100%. We also demonstrate the importance of the correct parameter settings in the PSO to guarantee good performance.

AB - Protein features are often complex, and they are challenging to classify. In identifying the most discriminatory features in protein sequences, we propose a new feature-selection strategy by integrating the multivariate filter and Particle Swarm Optimisation (PSO) algorithms. Experimental results, based on the number of reducts and classification accuracy, were analysed in both the filter and wrapper phases. For our dataset, the proposed method statistically significantly improves the obtained classification accuracy and reduces the number of feature subsets. In the filter phase, the accuracy is improved more than 4% in three out of four multivariate feature selection methods compared to a model without feature selection. In the second phase, the accuracy is increased from 97.51% to 100%. We also demonstrate the importance of the correct parameter settings in the PSO to guarantee good performance.

KW - Feature selection

KW - Multivariate filter

KW - Particle swarm optimisation

KW - Protein sequences

UR - http://www.scopus.com/inward/record.url?scp=79951488237&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79951488237&partnerID=8YFLogxK

U2 - 10.1109/SOCPAR.2010.5686158

DO - 10.1109/SOCPAR.2010.5686158

M3 - Conference contribution

AN - SCOPUS:79951488237

SN - 9781424478958

SP - 330

EP - 333

BT - Proceedings of the 2010 International Conference of Soft Computing and Pattern Recognition, SoCPaR 2010

ER -