Intelligent Data Analysis

Michael Berthold, David J. Hand

Mentioned 1

This second and revised edition contains a detailed introduction to the key classes of intelligent data analysis methods. The twelve coherently written chapters by leading experts provide complete coverage of the core issues. The first half of the book is devoted to the discussion of classical statistical issues. The following chapters concentrate on machine learning and artificial intelligence, rule induction methods, neural networks, fuzzy logic, and stochastic search methods. The book concludes with a chapter on visualization and an advanced overview of IDA processes.

More on

Mentioned in questions and answers.

In a follow-up to this answer I want to ask if any of you know any good (and more importantly easy to understand) tutorials and / or examples of data mining with the Weka toolkit.

I've been very interested in Data Mining ever since I've first heard of it and the things it can do, I've also have some experiments I'd like to do with some of my data and I've already bought four books and I found specially interesting the following two:

Intelligent Data Analysis Data Mining

The last one is written by the same authors of Weka and contains a lot of examples but still, I found it a little hard to understand the logic and specially the math. My math skills are currently very rough, I plan to go to the University this year and hopefully I'll learn and be able to better understand the math involved, but until then I want to gain some practice in Data Mining.

Is there any step-by-step tutorial with example data I can read to get me started with the Weka toolkit?

When it comes to "applied" data mining, for the starters, you may not need to think about math at all. Weka is product of a university machine-learning project and offers 100+ algorithms. Contrast that with Microsoft SQL server SSAS which offers nine algorithms -- and they do not even bother to explain the math.

They both offer association, clustering, attribute selection, some kind of neural network. So, the trick is to understand what you are trying to achieve, not necessarily the math below. Try reading about Microsoft algorithms (good documentation) and see if you can figure out principles that SSAS and Weka have in common -- this should help you focus on basic principles and get you started.

There is a list of a few Weka tutorials here.