The Scientific World Journal / 2014 / Article / Tab 6

Research Article

An Efficient Fitness Function in Genetic Algorithm Classifier for Landuse Recognition on Satellite Images

Table 6

K-HAT of each index varying with selection ways.

Selection wayIndex
K-HAT
DBIFCMIPASIDBFCMIAverage

Roulette wheel Selection0.360.340.240.480.34
Rank selection0.190.060.100.390.19
Standard deviation0.120.200.100.02

String length: 8, population: 30, crossover rate: 0.8, crossover way: two-point crossover, mutation rate: 0.003.