How does NRC work for feature extraction process?

[getamu]

Hello,
I have seen your program and I want to implement NRC emotion lexicon in English language. From what I can understand is that, we look for the emotion_value in the NRC emotion lexicon list and add the value for that emotions in a dictionary so as to get an entire score for the sentence. Do we need to binarize it later before we do machine learning or just apply machine learning on the dictionary itself?

Please Clarify.