International Journal of Advanced Research in Computer Science and Software Engineering (IJARCSSE)
Many organizations have large quantities of data collected in various application areas. Classification of data is a major issue which leads less efficiency and scalability. In this paper, the authors developed a new method for decision tree for classification of data using a data structure called Peano count tree (P-tree) which enhances the efficiency and scalability. They apply data smoothing and attribute relevance techniques along with a classifier. Experimental results show that the P-tree method is significantly faster than existing classification methods, making it the preferred method for mining on data to be classified.