DSE Data Science 2 — Questions and Answers
Question 1: Which technique is used to reduce the dimensionality of data while preserving as much variance as possible?
- Linear Discriminant Analysis
- Principal Component Analysis (Correct answer)
- K-Means Clustering
- Logistic Regression
Correct answer: Principal Component Analysis
PCA transforms data into orthogonal components ordered by the amount of variance they explain.
Question 2: In a confusion matrix, what does a False Negative represent?
- Model predicted positive and was correct
- Model predicted negative and was correct
- Model predicted positive but was wrong
- Model predicted negative but the actual was positive (Correct answer)
Correct answer: Model predicted negative but the actual was positive
A False Negative occurs when the model predicts the negative class but the true label is positive.
Question 3: What is the purpose of regularization in machine learning models?
- Speed up training time
- Increase model complexity
- Prevent overfitting by penalizing large coefficients (Correct answer)
- Normalize input features to zero mean
Correct answer: Prevent overfitting by penalizing large coefficients
Regularization adds a penalty term to the loss function to discourage overly complex models.
Question 4: Which of the following best describes the bias-variance tradeoff?
- High bias causes overfitting; high variance causes underfitting
- High bias causes underfitting; high variance causes overfitting (Correct answer)
- Both bias and variance decrease as model complexity increases
- Bias and variance are independent of model complexity
Correct answer: High bias causes underfitting; high variance causes overfitting
Simple models have high bias (underfitting) while complex models have high variance (overfitting).
Question 5: What does the term 'feature engineering' refer to in data science?
- Selecting the best algorithm for a task
- Creating or transforming input variables to improve model performance (Correct answer)
- Tuning hyperparameters of a model
- Evaluating model accuracy on test data
Correct answer: Creating or transforming input variables to improve model performance
Feature engineering involves creating new features or modifying existing ones to help the model learn better.
Question 6: In the context of gradient descent, what is the learning rate?
- The proportion of training data used each epoch
- The number of iterations to train the model
- A scalar that controls the step size during parameter updates (Correct answer)
- The ratio of correctly predicted samples
Correct answer: A scalar that controls the step size during parameter updates
The learning rate determines how large each update step is when minimizing the loss function.
Question 7: Which cross-validation strategy is most appropriate when data is time-ordered and temporal leakage must be avoided?
- K-Fold cross-validation
- Stratified K-Fold cross-validation
- Leave-One-Out cross-validation
- Time Series Split cross-validation (Correct answer)
Correct answer: Time Series Split cross-validation
Time Series Split ensures training folds always precede validation folds to prevent future data leaking into training.
DSE Data Science 2
Which technique is used to reduce the dimensionality of data while preserving as much variance as possible?