Machine Learning Classification 4

Question 1

What is the purpose of cross-validation in classifier evaluation?

Accepted Answer

To get a more reliable estimate of model performance by training and testing on multiple data splits

Answer

Cross-validation repeatedly splits data into train/test folds, averaging performance across folds to produce a less biased and more stable generalization estimate.

Question 2

Which boosting algorithm uses gradient descent in function space to minimize classification loss?

Accepted Answer

Gradient Boosting (GBM)

Answer

Gradient Boosting fits each new tree to the negative gradient of the loss function, effectively performing gradient descent to sequentially reduce errors.

Question 3

In a confusion matrix for binary classification, what does a 'False Positive' represent?

Accepted Answer

A negative sample incorrectly classified as positive

Answer

A false positive occurs when the model predicts the positive class but the true label is negative, also known as a Type I error.

Question 4

Which regularization technique in logistic regression adds the absolute value of coefficients as a penalty, promoting sparsity?

Accepted Answer

L1 (Lasso)

Answer

L1 regularization penalizes the sum of absolute values of weights, driving irrelevant feature coefficients exactly to zero and performing implicit feature selection.

Question 5

What is the F1 score?

Accepted Answer

The harmonic mean of precision and recall

Answer

The F1 score is the harmonic mean of precision and recall, giving a single metric that balances both and penalizes extreme imbalances between them.

Machine Learning Practice Test

Machine Learning Practice Test

Machine Learning Classification 4