This is a (so far small) collection of links and pointers to interesting software and methods to smart ways to mine data sets.
A talk by Steve Watts about data mining and visualisation. Also mentions SVM, see Essay2006 for some more links. Another way to find interesting things within data sets are Decision trees. Steve thinks all graduate students should read The visual display of quantitative information by Tufte.
Google videos on data mining: Statistical Aspects of Data Mining (Stats 202) Day 2, several videos in this lecture together with a website, notes, examples and so on.
Software
Orange, python scriptable
WEKA, machine learning workbench, written in Java, gentoo ebuild exists
Radviz and Polyviz help you to find out which variables are most interesting to look at. http://www.ailab.si:8088/supp/bi-vizrank/methods.htm claims to implement radviz