Methodology of the Heuristic Based Hybrid Clustering Technique for Pattern Classification and Recognition

In this paper we investigate the problem in different data sets to form similar objects into identical groups. Our technique is an unsupervised based algorithm. Unsupervised portion so high that the no input are given by user. Automatically judge the threshold applying threshold which is selected heuristic manner. It can also be resolve Singleton sets which can be identified in some special condition. Clustering is the clubbing of similar objects into identical groups, or more precisely, the partitioning of a data set into subsets (clusters), so that the data in each subset (ideally) share some common feature - often proximity according to some defined distance measure. Clustering is the clubbing of similar objects into identical groups, or more precisely, the partitioning of a data set into subsets (clusters), so that the data in each subset (ideally) share some common feature - often proximity according to some defined distance measure. The capability of recognizing and classifying patterns is one of the most fundamental characteristics of human intelligence. The primary goal of pattern recognition is supervised or unsupervised classification.