Fuzzy kernel clustering method based on improved Quantum-Behaved Particle Swarm Optimization algorithm

The fuzzy kernel clustering method (KFCM) is often sensitive to initial value and convergence to local optimum. In order to overcome these shortcomings, an improved Quantum-Behaved Particle Swarm Optimization (QPSO) algorithm was presented and used to optimize KFCM. In the improved QPSO which name as EQPSO (QPSO with Extrapolation), the particle take an extrapolation operator when it was worse than its previous generation particle in the evolution process. And then EQPSO-KFCM was put forward in which EQPSO is used to optimize the KFCM. The accuracy level of proposed algorithm EQPSO-KFCM was compared to those of KFCM and QPSO-KFCM on two test data set. The comparison indicates that EQPSO-KFCM clustering can be considered as a sufficiently accurate clustering method. In the end, EQPSO-KFCM is used in adaptability clustering of karst rocky desertification control patterns.