Does Balancing Classes Improve Classifier Performance?

http://ift.tt/1FHfMCY

It’s a folk theorem I sometimes hear from colleagues and clients: that you must balance the class prevalence before training a classifier. Certainly, I believe that classification tends to be easier when the classes are nearly balanced, especially when the class you are actually interested in is the rarer one. But I have always been … Continue reading Does Balancing Classes Improve Classifier Performance? → Related posts:

1. Don’t use correlation to track prediction performance
2. The Geometry of Classifiers
3. Can a classifier that never says “yes” be useful?

from R-bloggers http://ift.tt/xdvlrq