Confidence in hypotheses generated by data mining:

problem:

http://en.wikipedia.org/wiki/Texas_sharpshooter_fallacy

http://en.wikipedia.org/wiki/Multiple_comparisons

Solutions:

http://en.wikipedia.org/wiki/Familywise_error_rate

http://en.wikipedia.org/wiki/Bonferroni_correction

http://en.wikipedia.org/wiki/Tukey%27s_range_test

http://en.wikipedia.org/wiki/Scheffe_method

http://en.wikipedia.org/wiki/False_discovery_rate

http://en.wikipedia.org/wiki/False_coverage_rate

==============================================

http://www.jstor.org/stable/10.2307/2346101

http://www.jstor.org/stable/2346101

http://www.stat.cmu.edu/~genovese/talks/hannover1-04.pdf