MS-DS Master of Data science MS-DS Master of Data science Supervised Learning Algorithms Questions and Answers 2 — Questions and Answers
Question 1: Which supervised learning algorithm uses the concept of maximum margin to separate classes?
- Support Vector Machine (Correct answer)
- Naive Bayes
- K-Nearest Neighbors
- Logistic Regression
Correct answer: Support Vector Machine
SVMs find the hyperplane that maximizes the margin between the closest data points of different classes.
Question 2: In gradient boosting, what does each subsequent tree attempt to predict?
- The residual errors of the previous model (Correct answer)
- The original target variable
- The feature importances
- The learning rate
Correct answer: The residual errors of the previous model
Each new tree in gradient boosting is trained on the residual errors left by the ensemble of all previous trees.
Question 3: What is the primary assumption of the Naive Bayes classifier?
- Features are conditionally independent given the class label (Correct answer)
- The decision boundary is linear
- All features have equal variance
- The data is normally distributed
Correct answer: Features are conditionally independent given the class label
Naive Bayes assumes that all features are conditionally independent of each other given the class label.
Question 4: Which regularization technique in linear regression can shrink coefficients exactly to zero, enabling feature selection?
- L1 (Lasso) (Correct answer)
- L2 (Ridge)
- Elastic Net
- Dropout
Correct answer: L1 (Lasso)
L1 regularization (Lasso) penalizes the absolute value of coefficients, which can drive some coefficients to exactly zero.
Question 5: What problem does the kernel trick solve in Support Vector Machines?
- It allows SVMs to find non-linear decision boundaries without explicitly mapping to higher dimensions (Correct answer)
- It reduces overfitting by adding a penalty term
- It speeds up training on large datasets
- It handles missing values in the feature space
Correct answer: It allows SVMs to find non-linear decision boundaries without explicitly mapping to higher dimensions
The kernel trick computes dot products in a high-dimensional space without explicitly transforming the data, enabling non-linear classification.
Question 6: In a Random Forest, what technique is used to ensure diversity among individual decision trees?
- Bagging with random feature subsets at each split (Correct answer)
- Boosting with adaptive weights
- Stacking with a meta-learner
- Pruning with cross-validation
Correct answer: Bagging with random feature subsets at each split
Random Forest combines bootstrap aggregating (bagging) with random feature selection at each split to decorrelate the trees.
MS-DS Master of Data science MS-DS Master of Data science Supervised Learning Algorithms Questions and Answers 2
Which supervised learning algorithm uses the concept of maximum margin to separate classes?