An improved algorithm of decision tree for classifying large data set based on rainforest framework

Abstract

Data mining or Knowledge discovery is seen as an increasingly important tool by modern business to transform data into an informational advantage. Mining is a process of finding correlations among dozens of fields in large relational databases and extracts useful information that can be used to increase revenue, cuts costs, or both. Classification is a supervised machine learning procedure and an important issue in data mining. Of course Scalability is a major issue for mining large data set and it is unpractical that parsing the entire data set more than one time. This paper presents a more scalable decision tree algorithm which requires only one pass over the huge dataset for classification and also more efficient when compared with previous methods such as SLIQ, SPRINT, RainForest. It overcomes the drawback of Rainforest algorithm which majorly addresses scalability issue and requires a pass over the dataset in each level of decision tree construction. The experimental results show that our algorithm outperforms the RainForest algorithm for decision tree construction in time dimension. Moreover, our algorithm significantly reduces the sorting cost and hence the whole execution time.