The International Journal of Innovative Research in Computer and Communication Engineering
Many organizations having huge databases; the databases grow without limit at a rate of several million records per day. Mining these continuous data stream brings unique opportunities. VFDT use constant memory and constant time to builds decision trees per example. Using off the - shelf hardware, VFDT can to generate examples. The Hoeffding bounds output nearly similar to conventional learner. In this paper the authors introduce an effective algorithm for mining decision trees from massive data streams, based on the ultra-fast VFDT decision tree learner.