சுருக்கம்

Dynamic Learing Algorothm For Massive Data Streams

Ms.Manjusha Reddy, Prof. Rajesh Bharati

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 we introduce an effective algorithm for mining decision trees from massive data streams, based on the ultra-fast VFDT decision tree learner. Anthor algorithm defined as CVFDT,stays current while making the most of old data by growing an alternating sub tree whenever an old one becomes questionable ,and replacing the old with the new when the new becomes more accurate. CVFDT learns a model which is similar in accuracy to the one that would be learned by reapplying VFDT to a moving window of examples every time a new examples arrives, but with O(1)complexity per example ,as opposed to O(w),where w is the size of the window.