A comparison of particle swarm optimization algorithms in data clustering
详细信息
文摘
This thesis is an investigation into the use of Particle Swarm Optimization (PSO) techniques in data clustering. The PSO is an optimization technique based on swarm intelligence. The technique has been extended to data clustering. Several algorithms have been developed with some degree of success. In particular, three algorithms have been proposed. These include Dynamic Clustering using Particle Swarm Optimization (DCPSO), Exponential Particle Swarm Optimization (EPSO), and Particle Swarm-Like Agents Approach for Dynamically Adaptive Data Clustering (PSDC). This thesis attempts to compare these algorithms in the context of data clustering in terms of efficiency, convergence, and complexity. The comparison shows that each algorithm performs differently according to the size and dimensions of the datasets.