A HYBRID APPROACH USING C MEAN AND CART FOR CLASSIFICATION IN DATA MINING

Data Mining is a field of search and researches ofdata. Mining the data means fetching out a piece ofdata from a huge data block. The basic work in thedata mining can be categorized in two subsequentways. One is called classification and the other iscalled clustering. Although both refers to some kind ofsame region but still there are differences in both theterms. The classification of the data is only possible ifyou have modified and identified the clusters. In thepresented research paper, our aim is to find out themaximum number of clusters in a specified region byapplying the area searching algorithms. Classificationis always based on two things. a)The area which youchoose for the classification that is the cluster region.b)The kind of dataset which you are going to apply onthe selected region .To increase the accuracy of thesearching technique, any one would need to focus ontwo things . a)Whether the data set has been cauterizedin proper manner or not .b)If the clusters are defined ,whether they fit into the appropriate classified area ornot .