Confusion matrix normalize method

ipeirotis / Get-Another-Label

Quality control code for estimating the quality of the workers in crowdsourcing environments

70 stars 26 forks source link

Confusion matrix normalize method #34

Closed vic4ever closed 11 years ago

vic4ever commented 11 years ago

I see there are two normalizing methods for the confusion matrix: one uses Laplace smoothing, the other does not. Is there any difference between these ? I see that normalizeLaplacean is not used. Is there any problem if I use normalizeLaplacean ? The reason is that I want to get rid of the NaN thing.

ipeirotis commented 11 years ago

Feel free to change the code and use the Laplacean normalization. Just replace the calls to normalize() with normalizeLaplacean(). I made the decision to return NaN when it is impossible to compute a particular error rate or probability, instead of returning some arbitrary numbers set during the initialization.