I'm a developer, not too good at math, but I'm willing to learn fun stuff to do with data mining techniques.
I've looked at pragmatic books on the subject which gives me some ideas (and maybe introduces me rapidly some tools). I insist on the fact that I'm not a mathematician!
What do you advise?
I'm not qualified to judge the text compared to other works on the subject, but Witten and Frank's Data Mining: Practical Machine Learning Tools and Techniques keeps the mathematics to a minimum, and gives many practical examples.
Data Mining: Practical Machine Learning Tools and Techniques (Second Edition) by Witten & Frank worked for me.
The first part is about data mining theory and algorithms, the second how to use them in Weka data mining package.
From a NLP perspective, you should look at Introduction to Information Retrieval (Christopher D. Manning, Prabhakar Raghavan and Hinrich Schütze).
The prerequisites are : "Introductory courses in data structures and algorithms, in linear algebra, and in probability theory suffice as prerequisites for all twenty-one chapters."
All of those books are ok, but I think an even better book for developers is Programming Collective Intelligence by Toby Segaran.
I found Data Crunching to be very useful. Its not a Data Mining book but gives loads of practical techniques (with code) for searching, sorting, sifting, filtering data. Its like a lot of the Prag books: short and sweet, and all the more valuable for its brevity!
Yes I'd second that suggestion of Collective Intelligence in Action as a good intro - see the review Collective Intelligence in Action but also see Algorithms of the Intelligent Web which I think is a better book