subsample

A subsample is a sample (usually random) of a larger dataset. The dataset is itself is often a sample of a larger population., hence 'sub'.

Subsamples have several uses. A subsample can be used as a form of data reduction especially for early exploratory analysis, as it can be easier and faster to work with a smaller dataset. Subsamples are used in machine learning so that training and test data are different, a process known as holdout or jackknifing. Large numbers of random subsamples can also be used for empirical statstics in a process also known as bootstrapping.