Data Mining, the automatic extraction of implicit and potentially useful information from data, is increasingly used in commercial, scientific and other application areas. This book explains and explores the principal techniques of data mining: for classification, generation of association rules and clustering. It is written for readers without a strong background in mathematics or statistics and focuses on detailed examples and explanations of the algorithms given. This should prove of value to readers of all levels, from those whose use of data mining techniques will be only via commercial packages, right through to academic researchers. The book aims to help the general reader develop the necessary understanding to use commercial data mining packages discriminatingly, as well as enabling the advanced reader to understand or contribute to future technical advances in the field. Each chapter has practical exercises, and a glossary of technical terms is included.

This book explains the principal techniques of data mining: for classification, generation of association rules and clustering. It is written for readers without a strong background in mathematics or statistics and focuses on detailed examples and explanations of the algorithms given. This will benefit readers of all levels, from those who use data mining via commercial packages, right through to academic researchers. The book aims to help the general reader develop the necessary understanding to use commercial data mining packages, and to enable advanced readers to understand or contribute to future technical advances. Includes exercises and glossary.

Data for Data Mining.- to Classification: Näive Bayes and Nearest Neighbour.- Using Decision Trees for Classification.- Decision Tree Induction: Using Entropy for Attribute Selection.- Decision Tree Induction: Using Frequency Tables for Attribute Selection.- Estimating the Predictive Accuracy of a Classifier.- Continuous Attributes.- Avoiding Overfitting of Decision Trees.- More About Entropy.- Inducing Modular Rules for Classification.- Measuring the Performance of a Classifier.- Association Rule Mining I.- Association Rule Mining II.- Clustering.- Text Mining.