Principles of Data Mining

Abstract

This book is designed to be suitable for an introductory course at either undergraduate or masters level. It can be used as a textbook for a taught unit in a degree programme on potentially any of a wide range of subjects including Computer Science, Business Studies, Marketing, Artificial Intelligence, Bioinformatics and Forensic Science. It is also suitable for use as a self-study book for those in technical or management positions who wish to gain an understanding of the subject that goes beyond the superficial. It goes well beyond the generalities of many introductory books on Data Mining but—unlike many other books—you will not need a degree and/or considerable fluency in Mathematics to understand it. Mathematics is a language in which it is possible to express very complex and sophisticated ideas. Unfortunately it is a language in which 99% of the human race is not fluent, although many people have some basic knowledge of it from early experiences (not always pleasant ones) at school. The author is a former Mathematician who now prefers to communicate in plain English wherever possible and believes that a good example is worth a hundred mathematical symbols. One of the author’s aims in writing this book has been to eliminate mathematical formalism in the interests of clarity wherever possible. Unfortunately it has not been possible to bury mathematical notation entirely. A ‘refresher’ of everything you need to know to begin studying the book is given in Appendix A. It should be quite familiar to anyone who has studied Mathematics at school level. Everything else will be explained as we come to it. If you have difficulty following the notation in some places, you can usually safely ignore it, just concentrating on the results and the detailed examples given. For those who would like to pursue the mathematical underpinnings of Data Mining in greater depth, a number of additional texts are listed in Appendix C.