AUTOMATIC CLASSIFICATION OF E-MAIL MESSAGES BY MESSAGE TYPE

This article describes a system that automatically classifies E-mail m essages in the HUMANIST electronic discussion group into one of four c lasses, Question, Response, Announcement, or Administrative. A total o f 1,372 messages were analyzed. The automatic classification of a mess age was based on string matching between a message text and predefined string sets for each of the message types. The system's automated abi lity to accurately classify a message was compared against manually as signed codes. The Cohen's Kappa of .55 suggested that there was a stat istical agreement between the automatic and manually assigned codes.