Discriminant analysis for text genres
Karlgren and Cutting (94)
l Same text genre categories as Biber
l Simple count and average metrics
l Discriminant analysis (using SPSS software)
l 64% precision over four categories
•  Adverb
•
 Character
•
 Long word (> 6 chars)
•  Preposition
•
 2nd person pronoun
•
 “Therefore”
•
 1st person pronoun
•
 “Me”
•
 “I”
•
 Sentence
Text Box: Some count features
•  Words per sentence
•  Characters per word
•  Characters per sentence
•  Type / Token Ratio
Text Box: Other features