A Novel Dynamic Clustering Method by Integrating Marine Predators Algorithm and Particle Swarm Optimization Algorithm

Data clustering is the process of identifying natural groupings or clusters based on a certain similarity measure in muti-dimensional data. Aiming at dynamic problem where number cannot be determined advance, hybrid method marine predators algorithm (MPA) and particle swarm optimization (PSO) was proposed. The position update strategy PSO used to make up for lack MPA global searching. fixed-length coding with real deal variable length problem, unfeasible solution processing penalty function are adopted improve performance achieve simultaneous cluster centers. proposed MPA-PSO algorithm, MPA, Differential Evolution (DE) Spotted Hyena Optimizer (SHO), Lightning Searching Algorithm (LSA) Equilibrium (EO) carry out simulation experiments four artificial data sets six (Iris, Wine, Wisconsin breast cancer, Vowel, Seeds, Wdbc) UCI databases. Three indicators (the clusters, ARI Accuracy) evaluate results. experimental results show that can not only successfully find correct but also obtain stable most test problems.