Principles of Data Mining (Adaptive Computation and Machine Learning)

I bought this book because I wanted a relatively high level (not too high level, but high level enough to give me a good foundation in the theory and issues) to data mining. My goal was to first understand the theory and principles of data mining before getting into the technological and application specifics (e.g., how to use software such as Dataminer or R or Weka or SPSS Clementine etc.). This book has met my goals. Most chapters include abstract math/statistics that may be a little challenging for people who do not have a recent high level undergraduate statistics background. Actually I enjoyed the math/stats, and did not worry about going too deep into those portions. Trust me, the abstract concepts are not easy to grasp beyond a certain point, but they are EXTREMELY valuable. I am really glad that I was challeged. If you want another perspective or intro to data mining you may want to read some of the lecture notes of the "Machine Learning" course from MIT's online courseware - the courses are available for free on MIT's online courseware site. The lecture notes are even more abstract - they will make you appreciate this book. I highly recommend that anyone who wants to get an intro to data mining should first read this book. After reading this book the reader can read a book that explains a specific data mining software package such as "Intro to R" or "Data Mining: Practical Machine Learning Tools & Techniques" (by Witten and Frank, good if you want to learn Weka).