Data Science Data Science Supervised Learning Models Questions and Answers 3

Question 1

Which supervised learning method models the probability of a binary outcome using a logistic (sigmoid) function?

Accepted Answer

Logistic Regression

Answer

Linear Regression

Answer

Naive Bayes

Answer

K-Nearest Neighbors

Question 2

What is the bias-variance tradeoff in supervised learning?

Accepted Answer

The balance between a model's ability to fit training data closely and its ability to generalize to new data

Answer

The tradeoff between training speed and prediction accuracy

Answer

The balance between the number of features and the number of samples

Answer

The choice between supervised and unsupervised approaches

Question 3

In a k-Nearest Neighbors classifier, what effect does increasing the value of k typically have on the decision boundary?

Accepted Answer

The decision boundary becomes smoother and less sensitive to noise

Answer

The decision boundary becomes more complex and jagged

Answer

The model ignores the distance metric entirely

Answer

The model switches from classification to regression

Question 4

Which technique splits the dataset into multiple folds to train and validate a supervised model, ensuring every observation is used for both?

Accepted Answer

K-Fold Cross-Validation

Answer

Train-Test Split

Answer

Bootstrapping

Answer

Holdout Validation

Question 5

What assumption does Naive Bayes make about the relationship between features given the class label?

Accepted Answer

All features are conditionally independent given the class

Answer

Features must be normally distributed

Answer

Features must be standardized to zero mean

Answer

All features must be categorical

Question 6

What is the purpose of the cost function (loss function) in training a supervised learning model?

Accepted Answer

To quantify the difference between predicted and actual values so the model can be optimized

Answer

To determine how many features to include in the model

Answer

To assign weights to each training sample based on class frequency

Answer

To select which algorithm is best suited for the dataset