An Efficient Classification System Based on Binary Search Trees for Data Streams Mining

Abstract

Decision tree construction is a well-studied problem in data mining. Recently, there has been much interest in mining data streams. Domingos and Hulten have presented a one-pass algorithm for decision tree constructions. Their system using Hoeffding inequality to achieve a probabilistic bound on the accuracy of the tree constructed. In this paper, we revisit this problem and propose a decision tree classifier system that uses binary search trees to handle numerical attributes. The proposed system is based on the most successful VFDT, and it achieves excellent performance. The most relevant property of our system is an average large reduction in processing time, while keeps the same tree size and accuracy.