This paper presents the PLANMINE sequence mining algorithm to extract patte
rns of events that predict failures in databases: of plan executions. New t
echniques were needed because previous data mining algorithms were overwhel
med by the staggering number of very frequent, but entirely unpredictive pa
tterns that exist in the plan database. This paper combines several techniq
ues fur pruning out unpredictive and redundant patterns which reduce the si
ze of the returned rule set by more than three orders of magnitude. PLANMIN
E has also been fully integrated into two real-world planning systems. We e
xperimentally evaluate the rules discovered by PLANMINE, and show that they
are extremely useful for understanding and improving plans, as well as for
building monitors that raise alarms before failures happen.