Normally, for example, we talk about splitting datasets into training and test datasets. But. The splitting % per train and test sets happens in a subjective manner. Sometimes. The train is 60% or 70%, leaving the remaining for the test set. Sometimes we create validation sets and sometimes we don't. Sometimes we use accuracy and sometimes we use the ROC-AUC or F1. My question is simple. Is there standards (e.g. NIST, ISO) that dictates how the dataset and other model components need to be treated in order to meet a specific standard (compliance) in the financial, healthcare, or technology industry? Please provide citations not opinions.
Add a comment
|