Differential evolution and particle swarm optimisation in partitional clustering

摘要: Many partitional clustering algorithms based on genetic (GA) have been proposed to tackle the problem of finding optimal partition a data set. Very few studies considered alternative stochastic search heuristics other than GAs or simulated annealing. Two promising for numerical optimisation, which are hardly known outside field, particle swarm optimisation (PSO) and differential evolution (DE). The performance representative point approach is compared with PSO DE. empirical results show that DE clearly consistently superior hard problems, both respect precision as well robustness (reproducibility) results. Only simple sets, GA can obtain same quality Apart from performance, easy implement requires any parameter tuning substantial PSOs. Our study shows rather should receive primary attention in algorithms.