Performance Evaluation of Soft Computing using Clustering Techniques

Abstract Soft Computing can be considered as a tool to handle inexactness and uncertainty. The main concept of soft computing is to exploit the forbearance for inexactness, uncertainty, fractional truth and approximation to accomplish tractability, robustness and inexpensive solutions. It is used to solve real life problems full of uncomfortable features due to fractional, unclear, noisy and partial information. Clustering gets its name as a metaphor for the soft computing. Clustering aims to divide data groups into subgroups called clusters. Clustering is still searching for killer applications that not only takes advantage of its promise of "high performance and short development lead time". This paper focuses on the evaluation of the Genetic Algorithm concept of Soft Computing and measures its performance using various clustering techniques.