Data Mining: The Textbook

by Charu C. Aggarwal

Introduction to Data Mining.- Data Preparation.- Similarity and Distances.- Association Pattern Mining.- Association Pattern Advanced Concepts.- Cluster Analysis.- Cluster Advanced Concepts.- Outlier Analysis.- Outlier Advanced Concepts.- Data Classification.- Data Advanced Concepts.- Mining Data Streams.- Mining Text Data.- Mining Time-Series Data.- Mining Discrete Sequences.- Mining Spatial Data.- Mining Graph Data.- Mining Web Data.- Social Network Analysis.- Privacy-Preserving Data Mining.

Genres: Nonfiction, Computer Science, Computers, Programming, Technology, Reference

766 pages, Paperback

First published April 13, 2015

Reviews

[Terran M · Rating 5 out of 5]

Aggarwal deserves to be better known. This is an excellent survey of analytics and data mining models; it's unsuitable as a first book on the topic, but would be excellent as a 3rd or 4th. The great strength is the organized taxonomy in which techniques and subcomponents of techniques are presented. This also leads to some idiosyncracies, such as calling linear regression an advanced special case of classification, which is to put it mildly a minority view.

The strongest parts of the book are his descriptions of common components, such as distance functions, or meta algorithms; knowing about these concepts is often enough to implement them. The weakest parts are the description of individual algorithms, which are often too terse to follow unless you already understand what the algorithm or a related one does.

The best way to use this book is to first read one of the classics (e.g. Hastie and Tibshirani), and then read this book to fill in more options for how to alter and combine them. It also has significant coverage on association mining (i.e. recommender systems) which tend not to be treated elsewhere.

[Farzin]

[]