TMVA Interpretation

TMVA
1

Hello Everyone, Im working on TMVA classification using three classifiers BDT, Likelihood and MLP. im getting following plots and now im unable to interpret that what these plots tells. Cut efficiences and optimal cut value plot is for 5000 number of signal and background events. Also S/sqrt(S+B) curve is flat for likelihood. is there something wrong?

BDT
BDT578×750 83.1 KB

LH
LH588×708 63.8 KB

ML
ML592×717 78 KB

Regards

2

Hello @anwey ,
welcome to the ROOT forum and sorry for the high latency!

I believe our TMVA expert @moneta can help here.

Cheers,
Enrico

3

Hi,

I don’t see any problem with the plots above, they reflect the curves that you have below showing output signal and background distributions.

Lorenzo