Question: Suppose we are running a fraud classification model, with a training set of 10,000 records of which only 400 are fraudulent. How many fraudulent records
Suppose we are running a fraud classification model, with a training set of 10,000 records of which only 400 are fraudulent. How many fraudulent records need to be resampled if we would like the proportion of fraudulent records in the balanced data set to be 20%?
Step by Step Solution
3.44 Rating (154 Votes )
There are 3 Steps involved in it
To calculate the number of fraudulent records needed to be resampled we need to determine the size o... View full answer
Get step-by-step solutions from verified subject matter experts
