### Abstract

This paper presents the time complexity estimation and optimisation of the genetic algorithm clustering method. The tested feature in the clustering algorithm is the population limit function. For the purpose of the study, segmental kurtosis analysis was done on several segmented fatigue time series data, which are then represented in twodimensional heteroscaled datasets. These datasets are then clustered using the genetic algorithm clustering method and the runtime of the algorithm is measured against the number of iterations. Polynomial fitting is used on the runtime data to determine the time complexity of the algorithm. Analysis is repeated with the inclusion of the population limit in the clustering algorithm. The results of the analysis will be used to determine the significance of including the population limit function in the algorithm for optimal performance.

Original language | English |
---|---|

Pages (from-to) | 334-344 |

Number of pages | 11 |

Journal | WSEAS Transactions on Mathematics |

Volume | 9 |

Issue number | 5 |

Publication status | Published - May 2010 |

### Fingerprint

### Keywords

- Algorithm efficiency
- Big-O notation
- Clustering
- Fatigue damage
- Genetic algorithms
- Time complexity

### ASJC Scopus subject areas

- Mathematics(all)

### Cite this

*WSEAS Transactions on Mathematics*,

*9*(5), 334-344.

**Time complexity estimation and optimisation of the genetic algorithm clustering method.** / Mohd Nopiah, Zulkifli; Khairir, M. I.; Abdullah, Shahrum; Baharin, M. N.; Arifin, Azli.

Research output: Contribution to journal › Article

*WSEAS Transactions on Mathematics*, vol. 9, no. 5, pp. 334-344.

}

TY - JOUR

T1 - Time complexity estimation and optimisation of the genetic algorithm clustering method

AU - Mohd Nopiah, Zulkifli

AU - Khairir, M. I.

AU - Abdullah, Shahrum

AU - Baharin, M. N.

AU - Arifin, Azli

PY - 2010/5

Y1 - 2010/5

N2 - This paper presents the time complexity estimation and optimisation of the genetic algorithm clustering method. The tested feature in the clustering algorithm is the population limit function. For the purpose of the study, segmental kurtosis analysis was done on several segmented fatigue time series data, which are then represented in twodimensional heteroscaled datasets. These datasets are then clustered using the genetic algorithm clustering method and the runtime of the algorithm is measured against the number of iterations. Polynomial fitting is used on the runtime data to determine the time complexity of the algorithm. Analysis is repeated with the inclusion of the population limit in the clustering algorithm. The results of the analysis will be used to determine the significance of including the population limit function in the algorithm for optimal performance.

AB - This paper presents the time complexity estimation and optimisation of the genetic algorithm clustering method. The tested feature in the clustering algorithm is the population limit function. For the purpose of the study, segmental kurtosis analysis was done on several segmented fatigue time series data, which are then represented in twodimensional heteroscaled datasets. These datasets are then clustered using the genetic algorithm clustering method and the runtime of the algorithm is measured against the number of iterations. Polynomial fitting is used on the runtime data to determine the time complexity of the algorithm. Analysis is repeated with the inclusion of the population limit in the clustering algorithm. The results of the analysis will be used to determine the significance of including the population limit function in the algorithm for optimal performance.

KW - Algorithm efficiency

KW - Big-O notation

KW - Clustering

KW - Fatigue damage

KW - Genetic algorithms

KW - Time complexity

UR - http://www.scopus.com/inward/record.url?scp=77954771611&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77954771611&partnerID=8YFLogxK

M3 - Article

VL - 9

SP - 334

EP - 344

JO - WSEAS Transactions on Mathematics

JF - WSEAS Transactions on Mathematics

SN - 1109-2769

IS - 5

ER -