Thesis (BS Applied Mathematics) -- University of the Philippines Mindanao, 2008

Data clustering is a process of grouping together similar objects in bins. This project aims to find alternative method of clustering continuous data set using a hybrid type of algorithm. The two algorithms that I tried to hybrid are Particle Swarm Optimization and Tabu Search. These two algorithms are used in many fields of clustering. There are a lot of literatures about these two algorithms embedded in other existing algorithms. The results show that the hybrid method is a good alternative for the pure PSO algorithms in finding an optimum solution for iris data set and wine data set. The graphs show the comparison between the hybrid algorithm and its pure counterpart. However, the parameter settings may not be the optimum settings and maybe improved. Another comparison was made between PSO-TS and PSO-SA (Particle Swarm Optimization ? Simulated Annealing). The result shows that the hybrid method were possible alternative for the pure one depends on the preferred criteria of the researcher. The criteria used for this study are optimal quantization error and solution time.