WCCI Performance Prediction Challenge

The challenge is over, but a new challenge is on-going using the same datasets, check it out!

naive bayes f200 cv5

Submitted by Reference

naive bayes using feature selection to 200 features.
ber estimate using 5 fold cross validation

Dataset Balanced Error Test guess Guess error Test score Area Under Curve
Train Valid Test Train Valid Test
ada 0.3259 0.327 0.3243 0.3799 0.0556 0.3799 0.8827 0.875 0.8715
gina 0.238 0.2542 0.2414 0.2413 0.0001 0.2414 0.9071 0.9334 0.9024
hiva 0.2829 0.4017 0.3074 0.3009 0.0065 0.3114 0.8078 0.5905 0.7333
nova 0.0423 0.102 0.0965 0.1004 0.0039 0.0995 0.9993 0.9768 0.9825
sylva 0.0316 0.0289 0.0341 0.0335 0.0006 0.0345 0.9955 0.9959 0.9948
Overall 0.1841 0.2228 0.2007 0.2112 0.0134 0.2133 0.9185 0.8743 0.8969