NEW
Font size
WorksheetsMidterm_Quiz2_M7-8
Total questions: 20
Worksheet time: 10mins
What is the main goal of feature selection?
Increase data redundancy
Remove irrelevant and redundant features
Add more variables
Ignore model performance
Which problem often occurs when a dataset has too many features?
Underfitting
Missing values
Overfitting
Bias
PCA stands for:
Principal Classification Analysis
Principal Component Analysis
Primary Component Algorithm
Projection Coefficient Analysis
Which of the following is a filter method?
Correlation Coefficient Analysis
Recursive Feature Elimination
Decision Tree Importance
Lasso Regression
Which feature selection method evaluates subsets of features by testing model performance?
Filter
Wrapper
Embedded
Regularization
What type of dimensionality reduction is PCA?
Linear
Nonlinear
Hierarchical
Randomized
In PCA, new variables formed after transformation are called:
Eigenvectors
Principal Components
Latent Variables
Hidden Layers
The first principal component represents the:
Highest variance in data
Smallest variance
Noise in data
Average of features
Which method is most used for visualizing clusters in high-dimensional data?
PCA
t-SNE
LDA
Random Forest
What is the benefit of dimensionality reduction?
It increases processing time
It simplifies the dataset without losing much information
It makes models more complex
It removes outliers
Overfitting occurs when a model __________.
Ignores important features
Learns the noise in the training data
Has too few parameters
Has perfect test accuracy
Underfitting happens when a model __________.
Memorizes data patterns
Is too simple to capture data patterns
Overlearns the noise
Has too many layers
A model that performs well on training data but poorly on test data is likely __________.
Underfitted
Overfitted
Regularized
Balanced
Which of the following best describes underfitting?
The model performs poorly on both training and test data
The model performs perfectly on training data
The model learns the noise
The model is highly complex
What is the main cause of overfitting?
Too little data
Excessive model complexity
Poor feature scaling
Low regularization
What is the main cause of underfitting?
Too much regularization
Too simple model or missing important features
Overtraining
Data duplication
Which method adds penalties to control model complexity?
Regularization
Clustering
Normalization
Feature scaling
L1 regularization (Lasso) tends to __________.
Keep all coefficients equal
Remove unnecessary features
Increase coefficients
Add random noise
L2 regularization (Ridge) tends to __________.
Remove features completely
Ignore small features
Reduce coefficients without removing them
Overfit the data
Cross-validation helps by __________.
Adding more features
Increasing dataset size
Reducing noise manually
Testing model performance on different data subsets
