Predicting Failures in Event Sequences

In this paper we develop new techniques for predicting failures and monitoring in categorical event sequences. New techniques are needed because failures are rare and previous data mining algorithms were overwhelmed by the staggering number of very frequent, but entirely unpredictive patterns that exist in such databases. This paper combines several techniques for pruning out unpredictive and redundant patterns, which reduce the size of the returned rule set by more than three orders of magnitude. As a concrete application, we present PlanMine, an algorithm to extract patterns of events that predict failures in databases of plan executions. PlanMine has also been fully integrated into two real-world planning systems. We experimentally 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.