Data Science for Business: What You Need to Know about Data Mining and Data-Analytic Thinking

by Foster Provost

A common experimental methodology called cross-validation specifies a systematic way of splitting up a single dataset such that it generates multiple performance measures.