Building clusters with distributed features for text classification using KNN

Bulk data is generated in the era of Information Technology. If it is not stored in a properly systematic manner then the generated data cannot be reused. This is because navigation becomes if not impossible, certainly very difficult. So we classify the data before it is stored. Present paper explores the techniques to store the data in a supervised classification paradigm using distributed features. Initially Soft, hard and mixed Clusters are build based on the distributed features later the clusters are used to classify the documents based on the K-nearest neighbour classification algorithm.