Why Is Re-sampling Done?

Data Science, AI and ML
#1

Why Is Re-sampling Done?

#2
#3

We split our original data into training and testing sets. After finding suitable coefficients for model with the help of training set, we apply that model on testing set and find accuracy of the model. This is final accuracy before applying it to unknown data. And more is this final accuracy, higher is the hope to get accurate results on unknown data.

However, if we further divide training set into its training and testing subsets and then calculate final accuracy of that subset and repeatedly do this for many subsets then choose that coefficients(model) which gives us maximum accuracy among these subsets, and hope that this model will give maximum accuracy to final testing set.

The resampling is done to improve accuracy of model.

There are different ways of resampling of data like bootstrapping, cross-validation, repeated cross-validation etc.