STATISTICAL TECHNIQUES FOR TEXT CLASSIFICATION BASED ON WORD RECURRENCE INTERVALS
Matthew J. Berryman, Andrew Allison, Derek Abbott · Fluctuation and Noise Letters · 2003
We present a method for characterizing text based on a statistical analysis of word recurrence interval. This method can be used for extracting keywords from text, and also for comparing texts by an unknown author against a set of known authors. We also use these methods to comment on the controversial question of who wrote the letter to the Hebrews in the New Testament.