I am guessing the answer to this is probably not. The computational complexity of Fisher's Exact Test really explodes when you move beyond bi-grams. This is discussed a bit in one of my very early papers...
Significant Lexical Relationships (Pedersen, Kayaalp, & Bruce) - Appears in the Proceedings of the Thirteenth National Conference on Artificial Intelligence (AAAI-96), pp. 455-460, August 4-8, 1996, Portland, OR http://www.d.umn.edu/~tpederse/Pubs/aaai96-cmpl.pdf This doesn't mean it's impossible to implement, it's just a bit more complicated than extending some of the other measures to trigrams. I hope this helps! Cordially, Ted On Fri, Jan 30, 2009 at 4:28 AM, mercevg <merc...@yahoo.es> wrote:http://www.d.umn.edu/~tpederse/Pubs/aaai96-cmpl.pdf > Dear all, > > I would like to know if someone have tried to implement Fishers exact > test (left sided) as a trigram mesure. > > Thanks, > Mercè > > -- Ted Pedersen http://www.d.umn.edu/~tpederse