Recent Experiences with Data Mining in Aviation Safety
Earl Harris · 1999
While many data miners want to find “interesting” patterns, the definition of an interesting pattern is often unclear. This is a case study of data mining applied to the problem of improving aviation safety. To address the problem, we applied several methods to pertinent data; these methods achieved varying degrees of success. During the project, our notion of an interesting pattern evolved and our appreciation for a good working definition for interestingness grew.