Adaptive Stream Mining: Pattern Learning and Mining from Evolving Data Streams

This book focuses on the design of learning algorithms for evolving and time-changing data streams, specifically the adaptive sliding window algorithm (ADWIN) for change detection and value estimation and its use in predictive learning, clustering, and closed frequent tree mining from time-changing data streams. The first section introduces a framework for developing algorithms that can learn from data streams that change over time. It presents an ADWIN for detecting change and keeping statistics from a data stream updated. The second part of the book describes connected acyclic graphs, or 'trees,' from the point of view of closure-based mining, presenting efficient algorithms for subtree testing and for mining ordered and unordered frequent closed trees. The third section presents high-performance algorithms for mining rooted trees adaptively from data streams that change over time. There is also a general methodology presented for identifying closed patterns in a data stream, with examples of three different types. Bifet teaches computer science at the University of Waikato, New Zealand. Annotation c2010 Book News, Inc., Portland, OR (booknews.com)