This book is a significant contribution to the subject of mining time-changing data streams and addresses the design of learning algorithms for this purpose. It introduces new contributions on several different aspects of the problem, identifying research opportunities and increasing the scope for applications. It also includes an in-depth study of stream mining and a theoretical analysis of proposed methods and algorithms.
The first section is concerned with the use of an adaptive sliding window algorithm (ADWIN). Since this has rigorous performance guarantees, using it in place of counters or accumulators, it offers the possibility of extending such guarantees to learning and mining algorithms not initially designed for drifting data. Testing with several methods, including Naïve Bayes, clustering, decision trees and ensemble methods, is discussed as well. The second part of the book describes a formal study of 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. Lastly, a general methodology to identify closed patterns in a data stream is outlined. This is applied to develop an incremental method, a sliding-window based method, and a method that mines closed trees adaptively from data streams. These are used to introduce classification methods for tree data streams.
Knowledge Discovery from data streams is one of the most relevant challenges that we face today. Data mining algorithms for analyzing static data sets, assuming stationary distributions, unlimited memory, and generating static models are almost obsolete for the real challenging problems we are faced nowadays.
Albert's research is not in the established data mining. He does steps ahead, focusing his research in cutting-the-edge topics, methods and algorithms. His work focuses on evolving data, going far way from iid assumption. Models are not static, training sets are not fixed. Albert's approach tries to capture the dynamics of patterns in the problem under study: how patterns evolve, grow, decrease and die.
The book focus on the design of learning algorithms for evolving and time-changing data streams. The basic building block is the adaptive sliding window algorithm, ADWIN, for change detection and value estimation with strong theoretical guarantees. The book presents evidence of the advantages of the framework in predictive learning, clustering, and closed frequent tree mining from time-changing data streams. The work opens new research opportunities and a large space for applications. The work presents an in-depth study in streaming mining. The author carefully presents theoretical analysis of the proposed methods and algorithms. It introduces new contributions on several different aspects of the problem. Overall this is an interesting dissertation that makes significant contributions to mining time-changing streams.
It is not a surprise, the author's good record of publications in top-level conferences. The ideas come from and reflect a world in movement: a pleasure to read. I'm sure you will enjoy it.
IOS Press, Inc.
6751 Tepper Drive
Clifton, VA 20124
Tel.: +1 703 830 6300
Fax: +1 703 830 2300 firstname.lastname@example.org
(Corporate matters and books only) IOS Press c/o Accucoms US, Inc.
For North America Sales and Customer Service
West Point Commons
Lansdale PA 19446
Tel.: +1 866 855 8967
Fax: +1 215 660 5042 email@example.com