Data for machine learning · Question 59 of 60
Which other resampling methods for scarce data exist besides k-fold cross-validation?
2 answers are correct. Hard
Show answer
Answer
- C Leave-one-out cross-validation, in which k equals the number of samples
- D Bootstrap techniques
Explanation
When data is scarce, further resampling methods help: leave-one-out cross-validation, in which each individual data point is used once for checking (k equals the number of data points), and bootstrapping. Back-to-back and metamorphic tests check models; they are not resampling methods.