Voluntary error reporting systems are an important part of safety improvement programs, but difficulty in analyzing error reports has limited their utility. This study described the development of a machine learning algorithm to analyze free-text data in incident reports. The algorithm proved to be accurate in classifying events when compared to manual review.