[Corpora-List] Significance tests for n-fold cross validation

[Gemma Boleda]

Dear Emad,

I use the "corrected resampled t-test" by Nadeau & Bengio for comparing 
results obtained with different features (also with different 
algorithms) on a classification task:

Nadeau, Claude and Yoshua Bengio. 2003. Inference for the generalization 
error. Machine Learning, 52(3):239–281, September.

Best,
Gemma.

-- 
Gemma Boleda
Post-doc researcher
TALP - GPLN
Universitat Politècnica de Catalunya
http://www.lsi.upc.edu/~gboleda


_______________________________________________
UNSUBSCRIBE from this page: http://mailman.uib.no/options/corpora
Corpora mailing list
Corpora at uib.no
http://mailman.uib.no/listinfo/corpora