A data analyst trained a machine learning model to classify whether a transaction is fraudulent or not. The model had a training score of 89% and a test score of 92% percent. It was suggested that the high score on the test set may be due to the fact that the test split was easy to predict on. The analyst was then advised to use cross validation so that he gets metrics on different and random splits. Which of the following is/are the possible algorithms that can be used?- i. KFold cross validation- ii. Shuffle Split Cross validation- iii. Grid search cross validation- iv. Linear regressioni. onlyi. and ii onlyiii. and iv onlyiii. onlyiv. onlyii. only

What is the purpose of the k-fold cross-validation technique in machine learning?a.To evaluate a model's performance on a separate test dataset.b.To reduce the risk of overfitting by training and testing a model on different data subsets.c.To speed up the training process by using parallel computing.d.To partition the dataset into k equal subsets for training and testing.

Cross-validation is used to: Test a model on new data Train a model on multiple datasets Evaluate model performance on a held-out test set Simulate the training process

What is the purpose of cross-validation in machine learning?(1 Point)To evaluate the performance of a model on a held-out test setTo evaluate the performance of a model on different subsets of the dataTo compare the performance of different modelsTo tune the hyperparameters of a model

Regarding splitting datasets into training, validation, and test partitions, which ofthe following statements is true, if any?(i) The validation set is used multiple times to choose the best value forhyperparameters.(ii) The test set is used only once to determine the performance on unseen data.(iii) Improving performance on the validation set always improves performance onthe test set.

What is the main characteristic of Shuffle Split Cross-Validation?Review LaterIt preserves the class distribution within each foldIt uses historical data for training and recent data for validationIt creates random train/validation splits with controlled proportionsIt ensures that samples belonging to the same group are kept together