Outlier Analysis

by Charu C. Aggarwal

With the increasing advances in hardware technology for data collection, and advances in software technology (databases) for data organization, computer scientists have increasingly participated in the latest advancements of the outlier analysis field. Computer scientists, specifically, approach this field based on their practical experiences in managing large amounts of data, and with far fewer assumptions– the data can be of any type, structured or unstructured, and may be extremely large. Outlier Analysis  is a comprehensive exposition, as understood by data mining experts, statisticians and computer scientists. The book has been organized carefully, and emphasis was placed on simplifying the content, so that students and practitioners can also benefit. Chapters will typically cover one of three areas: methods and techniques  commonly used in outlier analysis, such as linear methods, proximity-based methods, subspace methods, and supervised methods; data  domains, such as, text, categorical, mixed-attribute, time-series, streaming, discrete sequence, spatial and network data; and key applications of these methods as applied to diverse domains such as  credit card fraud detection, intrusion detection, medical diagnosis, earth science, web log analytics, and social network analysis are covered.

Genres: Mathematics, Science

461 pages, Hardcover

First published January 1, 2013

Reviews

[Terran M · Rating 5 out of 5]

This is an excellent, clearly-written book on outlier analysis and its cousin, anomaly detection. In my opinion the author uses exactly the right amount of equations, coupled with good prose, to clearly convey the concepts and provide a coherent framework to attach them to. If you liked Introduction to Statistical Learning by Jame et al, I think you will also like this book.

Note that although it is well written, this is not an introductory book. To decide if you are ready for it, I recommend downloading and reading the first chapter from the author's website. Look at the exercises at the end; if you can do them easily, you're ready for this book. If you have any difficulty at all, go read or review Hastie and Tibshirani first and then come back.

[Vinicius Catelani · Rating 5 out of 5]

It is an amazing book to learn more about outlier analysis and detection, it is an intuitive book, it doesn't have much math but it uses the right examples and amount of math necessary to understand each possible application of outlier analysis and detection. However, the book is not for beginners, it is necessary a previous, but not so deep, knowledge of linear algebra and calculus so you can follow the explanations with ease.
Two books I recommend reading before starting this one are "APEX Calculus", by Gregory Hartman, and "Mathematics for Machine Learning", by Marc Peter Deisenroth, they both will give you enough basis to follow all the explanations about outlier analysis and detection.

[Raoul · Rating 4 out of 5]

Clear and comprehensive. The section on neural networks could be expanded.