Hands-On Artificial Intelligence for Cybersecurity

K-folds cross validation pros and cons

K-folds cross validation has several advantages, including the following:

• It enables all the available data to be used, both for training and testing goals
• The specific composition of the individual folds is irrelevant since each fold is used at most once for training, and once for testing
• We can increase the number of folds by increasing the k value, thereby increasing the size of the training dataset to reduce the Bias component of the generalization error

In terms of disadvantages, we must stress that k-folds cross validation considers the order that characterizes our original sample dataset to be irrelevant. In case the order of the data constitutes relevant information (as in the case of time ...