... 785–792,Sydney, July 2006.c2006 Association for Computational Linguistics You Can t Beat Frequency (Unless You Use Linguistic Knowledge) –A Qualitative Evaluation of Association Measures ... all, these kinds of commonly used sta-tistical significance tests may not provide the rightmachinery in the first place. By design, they arerather limited (or focused) in their scope in thatthey ... recognition (ATR), we used a biomedicalterm candidate set put forth by Wermter and Hahn(2005), who, after shallow syntactic analysis, ex-tracted 31,017 trigram term candidates occurringat...