I'm working on an experiment using KDD'99 cupset I have 42 features. the paper I 'm comparing with concludes that 3 features with precision ..% ok are the best subset to identify the attack X. In my experiment, I applied 4 different classifier through the PCA. How to compare between them in order to conclude the number of features used in my experiment ? how to explain my features in order to say that n features gives higher preciosion.

  • $\begingroup$ Welcome to DataScienceSE. Can you please edit the question, it's not clear exactly what you are doing and what you need help with. $\endgroup$
    – Erwan
    Dec 21, 2020 at 20:25