Selection of initial points, the number of clusters and finding proper clusters centers are\nstill the main challenge in clustering processes. In this paper, we suggest genetic algorithm\nbased method which searches several solution spaces simultaneously. The solution spaces are\npopulation groups consisting of elements with similar structure. Elements in a group have the\nsame size, while elements in different groups are of different sizes. The proposed algorithm\nprocesses the population in groups of chromosomes with one gene, two genes to k genes.\nThese genes hold corresponding information about the cluster centers. In the proposed method,\nthe crossover and mutation operators can accept parents with different sizes; this can lead to\nversatility in population and information transfer among sub-populations. We implemented the\nproposed method and evaluated its performance against some random datasets and the Ruspini\ndataset as well. The experimental results show that the proposed method could effectively\ndetermine the appropriate number of clusters and recognize their centers. Overall this research\nimplies that using heterogeneous population in the genetic algorithm can lead to better results.
Loading....