Nikravesh and Takagi identify the use of fuzzy or neuro-fuzzy clusters to classify without any supervision based upon conventional learning technique and Genetic & Reinforcement learning. Such algorithms have not been widely explored for clustering of documents although a research at indicates that the Fuzzy c-Means algorithm performs at-par with the traditional agglomerative-hierarchical-clustering methods. Improvment in search & retrieval efficiencies is achieved in information retrieval by application of clustering techniques. Cluster hypothesis supports the use of clustering for information retrival . It is based on the assumption that document which are resultant of a specific query would be more similar to one another in comparison to other unrelated document thus these relevant document are more likely to be bunched with each other. To browse a large collection of documents, clustering is proposed as a tool which can also be used as a tool for organization of search results on the web and meaningful groups after retrieval.