Clustering is the unsupervised learning problem. Better Clustering improves accuracy of search results and  helps to reduce the retrieval time. Clustering dispersion known as entropy which is the disorderness that occur after  retrieving search result. It can be reduced by combining clustering algorithm with the classifier. Clustering with weighted  k-mean results in unlabelled data. This paper present a clustering algorithm called Minkowski Weighted K-Means. This algorithm automatically calculates feature weights for each cluster and uses the Minkowski metric (Lp) Unlabelled data can be labeled by using neural network and support vector machines. A neural network is an interconnected group of nodes, for classifying data whereas SVM is the classification function to distinguish between members of the two classes in the training data. For classification we use neural networks and SVM as they can recognize the patterns. The whole work is taken place in the Matlab.7 environment.