188
Y. Zhang et al.
Fig. 17.1 The flow of
clustering analysis based on
improved k-means algorithm
center is modified to the number of the cluster center to ensure that the labels of basic
clusters are consistent.
17.3.2 Genetic Operation and Fitness Function Selection
The main genetic operations in unsupervised clustering based on improved genetic
algorithm are crossover, mutation and selection operator selection. In this paper,
single point crossing, random variation and roulette according to individual fitness
are used to carry out genetic operation. Fitness is usually used to measure the degree of
excellence of individual in the group to reach the optimal solution in the optimization
calculation. In this paper, the fitness function is constructed by formula (17.1). The
smaller the J value in formula (17.1) is, the better the clustering result is. Therefore,
the following fitness function is selected:
f =
c
1 + J
(17.1)
where c is a constant and J is a criterion function.
Y. Zhang et al.
Fig. 17.1 The flow of
clustering analysis based on
improved k-means algorithm
center is modified to the number of the cluster center to ensure that the labels of basic
clusters are consistent.
17.3.2 Genetic Operation and Fitness Function Selection
The main genetic operations in unsupervised clustering based on improved genetic
algorithm are crossover, mutation and selection operator selection. In this paper,
single point crossing, random variation and roulette according to individual fitness
are used to carry out genetic operation. Fitness is usually used to measure the degree of
excellence of individual in the group to reach the optimal solution in the optimization
calculation. In this paper, the fitness function is constructed by formula (17.1). The
smaller the J value in formula (17.1) is, the better the clustering result is. Therefore,
the following fitness function is selected:
f =
c
1 + J
(17.1)
where c is a constant and J is a criterion function.