📢 Too many exams? Don’t know which one suits you best? Book Your Free Expert 👉 call Now!
  • google app store apple app store
    • ✖

      Question

      'Cross-validation' means:

      A Validating that training and test datasets come from the same distribution by cross-checking the source of both datasets Correct Answer Incorrect Answer
      B Splitting data into k subsets/folds, training the model k times using a different fold as validation and the rest for training Correct Answer Incorrect Answer
      C Comparing two different machine learning algorithms on the same dataset and cross-checking the results obtained Correct Answer Incorrect Answer
      D Validating model predictions against actual outcomes after production deployment Correct Answer Incorrect Answer
      E A process of manually reviewing model predictions for quality assurance by cross-checking the model functions Correct Answer Incorrect Answer

      Solution

      k-Fold Cross-Validation (typically k=5 or 10) overcomes the limitation of a single train/test split, which can be lucky or unlucky based on which data ends up in each set. Each data point appears in the validation set exactly once. It provides k performance estimates whose average and standard deviation give a reliable picture of model performance and stability. Stratified k-Fold maintains class proportions in each fold which is important for imbalanced banking datasets as fraud cases are rare. Leave-One-Out (LOO) is an extreme case where k = n (dataset size).

      Practice Next
      More Data Analytics Languages Questions

      Relevant for Exams:

      ask-question