Is $F_1$ Score Suboptimal for Cybersecurity Models? Introducing $C_{score}$, a Cost-Aware Alternative for Model Assessment

Manish Marwah, Asad Narayanan, Stephan Jou +2

The cost of errors related to machine learning classifiers, namely, false positives and false negatives, are not equal and are application dependent. For example, in cybersecurity…