WCCI Performance Prediction Challenge

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

linearSVC_Ensemble_CV

Submitted by Reference

ensemble of linear svM
BER estimation by 5 fold CV

CLOP definition:
for k=1:3
if strcmp(data_name, 'nova')
base_model{k}=svc(['shrinkage=' num2str(10^-(k-1))]); % Standardizing nova poses memory problems
else
base_model{k}=chain({standardize, svc(['shrinkage=' num2str(10^-(k-1))]) });
end
end
my_model=ensemble(base_model, 'signed_output=1')

Dataset Balanced Error Test guess Guess error Test score Area Under Curve
Train Valid Test Train Valid Test
ada 0.2201 0.2187 0.2368 0.2206 0.0162 0.253 0.8092 0.8055 0.7875
gina 0.1306 0.1277 0.1406 0.1503 0.0097 0.1503 0.8693 0.8723 0.8595
hiva 0.2488 0.3971 0.3162 0.3169 0.0007 0.3163 0.7512 0.6029 0.6838
nova 0 0.112 0.0967 0.1113 0.0146 0.1113 1 0.9073 0.9153
sylva 0.0859 0.0835 0.0847 0.0842 0.0005 0.085 0.9141 0.9165 0.9153
Overall 0.1371 0.1878 0.175 0.1767 0.0084 0.1832 0.8688 0.8209 0.8323