A study on fuzzy and particle swarm optimization algorithms and their applications to clustering problems

Data mining refers to the finding of relevant and useful information from the databases. Clustering is one of the important data mining tasks. It is the process of grouping objects into clusters such that the objects from the same cluster are similar and objects from different clusters are dissimilar. Many algorithms have been proposed in the literature. Fuzzy c-means algorithm is one of the popular clustering techniques. However, it will get easily struck at local minima. Recently, the use of global optimization technique such as particle swarm optimization has emerged in clustering field. In this paper, we present both fuzzy c-means and particle swarm optimization algorithms to solve the clustering problems. A brief review of applications is also described. The fuzzy c-means algorithm is experimented with different distance measures like Euclidian, Angular separation and Canberra for well known real-world data sets.