MCQs

Disclaimer:

Educational Use Only:

The website links provided are to be used strictly for educational and knowledge-gaining purposes only.

No Misuse:

Do not misuse the links or the information contained therein for any commercial, illegal, or unethical activities.

Respect for Websites and Businesses:

Respect the intellectual property and business interests of the websites and their owners. Do not attempt to alter, copy, or redistribute content without proper authorization.

Acknowledge Value of Knowledge:

Recognize and value the knowledge and resources shared by these websites. Use the information responsibly and ethically.

No Endorsement:

The provision of these links does not imply endorsement of the linked sites, their content, or their products and services. Users are responsible for evaluating the suitability and accuracy of the information.

User Responsibility:

By accessing these links, you agree to use the information at your own risk and to comply with all applicable laws and website terms of use.

Overall MCQs for Reference

What is the primary goal of Principal Component Analysis (PCA)?
a. To increase the number of features
b. To reduce the number of features while preserving data variance
c. To visualize data as images
d. To encrypt the dataset
Answer: b. To reduce the number of features while preserving data varianceibm+2
Which mathematical concept is at the core of PCA?
a. Linear regression
b. Support vector machines
c. Eigenvectors and eigenvalues
d. K-means clustering
Answer: c. Eigenvectors and eigenvaluesgeeksforgeeks+2
PCA transforms correlated variables into which new set?
a. Principal components
b. Random forests
c. Clusters
d. Trees
Answer: a. Principal componentssimplilearn+2
What type of learning does PCA represent in machine learning?
a. Supervised
b. Unsupervised
c. Semi-supervised
d. Reinforcement
Answer: b. Unsupervisedkeboola+1
Why is dimensionality reduction important in AI?
a. It increases storage costs
b. It makes computation faster and models less likely to overfit
c. It slows down visualization
d. It hides data patterns
Answer: b. It makes computation faster and models less likely to overfitappliedaicourse+1
What is the first step in a typical PCA process?
a. Data normalization
b. Covariance matrix calculation
c. Feature extraction
d. Data compression
Answer: a. Data normalizationpca4ds.github+2
The covariance matrix in PCA helps to measure?
a. The frequency of sampling
b. The relationships between variables
c. The number of clusters
d. The entropy of data
Answer: b. The relationships between variablesgeeksforgeeks+1
In PCA, how are principal components typically ordered?
a. Alphabetically
b. By descending eigenvalues
c. By mean value
d. By the number of features
Answer: b. By descending eigenvaluesbuiltin+1
What is the primary application of Kernel PCA?
a. Detecting linear relationships
b. Capturing non-linear data structures
c. Increasing computational cost
d. Recognizing hand-written digits
Answer: b. Capturing non-linear data structuresbuiltin
What advantage does Sparse PCA offer?
a. Uses all available features
b. Enhances interpretability by producing sparse loadings
c. Duplicates original data
d. Ignores eigenvalues
Answer: b. Enhances interpretability by producing sparse loadingsbuiltin
What is the Kaiser criterion in PCA?
a. Selecting components with eigenvalues less than 1
b. Choosing components with eigenvalues greater than 1
c. Picking random components
d. Using all available components
Answer: b. Choosing components with eigenvalues greater than 1scirp
Which step follows the calculation of the covariance matrix in PCA?
a. Clustering
b. Computation of eigenvectors and eigenvalues
c. Model training
d. Data encryption
Answer: b. Computation of eigenvectors and eigenvaluespca4ds.github+1
What percentage of data variance is typically retained by the first few principal components in a well-constructed PCA?
a. Less than 20%
b. About 50%
c. Over 90%
d. 100%
Answer: c. Over 90%scirp+1
What is a scree plot in the context of PCA?
a. A plot of feature counts
b. A plot of eigenvalues in descending order
c. A plot of mean vectors
d. A plot of raw data
Answer: b. A plot of eigenvalues in descending orderscirp
Which kind of datasets benefit most from PCA?
a. Low-dimensional datasets
b. High-dimensional datasets
c. Datasets with unique feature values
d. Binary datasets
Answer: b. High-dimensional datasetspmc.ncbi.nlm.nih+1
What does data compression mean in PCA?
a. Combining datasets
b. Reducing the number of dimensions while keeping essential data characteristics
c. Adding random noise
d. Encrypting data
Answer: b. Reducing the number of dimensions while keeping essential data characteristicssimplilearn+1
Which field is PCA NOT usually applied to?
a. Drug discovery
b. Psychology
c. Meteorology
d. Hardware manufacturing
Answer: d. Hardware manufacturingsciencedirect
Principal components are always:
a. Correlated
b. Uncorrelated
c. Noisy
d. Weighted by sample variance
Answer: b. Uncorrelatedwikipedia+1
What pre-processing step is usually necessary before applying PCA?
a. Data normalization or standardization
b. Feature selection
c. Random sampling
d. Data encryption
Answer: a. Data normalization or standardizationappliedaicourse+2
How does PCA help with noise reduction?
a. By including all low-variance features
b. By filtering out components with low variance
c. By increasing data variance
d. By scrambling feature values
Answer: b. By filtering out components with low variancesimplilearn+1
In PCA, what do eigenvalues represent?
a. The direction of maximum variance
b. The amount of variance explained by each principal component
c. The mean of each variable
d. The frequency of sampling
Answer: b. The amount of variance explained by each principal componentpca4ds.github+2
What is the typical visualization made possible by PCA for high-dimensional data?
a. 1D scatter plots
b. 2D or 3D scatter plots of principal components
c. Histograms of raw features
d. Decision trees
Answer: b. 2D or 3D scatter plots of principal componentsappliedaicourse+1
Which algorithm is commonly used to solve for principal components?
a. Gradient descent
b. Singular value decomposition (SVD)
c. Decision tree algorithm
d. Random forest
Answer: b. Singular value decomposition (SVD)towardsdatascience+1
What criterion helps decide how many components to keep in PCA?
a. Minimum error criterion
b. Cumulative explained variance threshold
c. Maximum likelihood criterion
d. Frequency criterion
Answer: b. Cumulative explained variance thresholdbuiltin+1
In PCA, which of the following is true about the first principal component?
a. It accounts for the least variance in the data
b. It is always discarded
c. It accounts for the maximum variance in the data
d. It is always normalized
Answer: c. It accounts for the maximum variance in the datageeksforgeeks+2
What type of dataset is most suitable for PCA?
a. Data with highly correlated features
b. Data with independent features
c. Data with class labels
d. Very small datasets
Answer: a. Data with highly correlated featureswikipedia+1
What is an example of PCA application in image processing?
a. Compression of image data into fewer principal components
b. Image encryption
c. Edge detection
d. Pixel scaling
Answer: a. Compression of image data into fewer principal componentssimplilearn+1
What is meant by “loadings” in PCA?
a. The amount of variance each principal component explains
b. The coefficients of original variables in each principal component
c. The noise in each component
d. The sample frequency
Answer: b. The coefficients of original variables in each principal componentwikipedia+1
What type of PCA formulation maximizes the variance retained?
a. Maximum variance formulation
b. Minimum error formulation
c. Maximum likelihood formulation
d. Support vector PCA
Answer: a. Maximum variance formulationtowardsdatascience
PCA can be problematic if the dataset has:
a. Features with vastly different scales and not standardized
b. Sufficiently normalized features
c. Only binary features
d. Features with equal mean
Answer: a. Features with vastly different scales and not standardizedgeeksforgeeks+1
Which of the following is NOT a benefit of PCA?
a. Data visualization
b. Noise reduction
c. Increasing data dimension
d. Feature extraction
Answer: c. Increasing data dimensionappliedaicourse+1
Which method can be used to decide the optimal number of principal components?
a. Scree plot
b. Silhouette plot
c. ROC curve
d. K-means clustering
Answer: a. Scree plotscirp
When applying PCA to time-series data, what must be considered?
a. Non-independence of samples
b. The data mean
c. Eigenvalue normalization
d. Random permutation
Answer: a. Non-independence of samplespmc.ncbi.nlm.nih
What is the role of the covariance matrix in PCA?
a. It determines the clusters in data
b. It shows the linear relationship among features
c. It provides the mean value per feature
d. It outputs class labels
Answer: b. It shows the linear relationship among featurespca4ds.github+1
Which Principal Component corresponds to the highest eigenvalue?
a. Farthest from the origin
b. First principal component
c. Last principal component
d. Arbitrary principal component
Answer: b. First principal componentgeeksforgeeks+1
Which kind of features are most likely eliminated during PCA?
a. Features with highest variance
b. Features with lowest variance
c. Features with maximum correlation
d. Features with maximum entropy
Answer: b. Features with lowest varianceknowledge.dataiku+2
What is the minimum number of principal components needed to reconstruct the original data exactly?
a. One
b. Equal to the number of original features
c. Half the number of features
d. Zero
Answer: b. Equal to the number of original featureswikipedia+1
In PCA, after transformation, the data is represented as a linear combination of:
a. Random numbers
b. Principal components
c. Class labels
d. Gaussian noise
Answer: b. Principal componentssimplilearn+2
What error can be minimized by using the minimum error formulation in PCA?
a. Classification error
b. Reconstruction error
c. Clustering error
d. Prediction error
Answer: b. Reconstruction errortowardsdatascience
What happens if only one principal component is retained in PCA?
a. All compositional information is preserved
b. Data is reduced to its maximal variance direction
c. Data entropy increases
d. All noise remains
Answer: b. Data is reduced to its maximal variance directionbuiltin+1
Which transformation does PCA use for feature decorrelation?
a. Orthogonal transformation
b. Affine transformation
c. Random permutation
d. Exponential transformation
Answer: a. Orthogonal transformationknowledge.dataiku+2
PCA is widely used in what data pre-processing step?
a. Feature engineering for clustering and classification
b. Hyperparameter tuning
c. Data cleaning
d. Data encryption
Answer: a. Feature engineering for clustering and classificationappliedaicourse+1
What is the effect of PCA on overfitting in models?
a. It increases the risk
b. It reduces the risk
c. It has no effect
d. It causes underfitting
Answer: b. It reduces the risksimplilearn+1
What dimension will a dataset reduce to after PCA if only the first principal component is used?
a. Two dimensions
b. One dimension
c. Three dimensions
d. Equal to the original
Answer: b. One dimensiongeeksforgeeks+1
In PCA, which component has the least information about the data?
a. First principal component
b. Last principal component
c. Component with the highest eigenvalue
d. Component with the highest loading
Answer: b. Last principal componentbuiltin
What kind of principal components does PCA generate?
a. Linear combinations of original features
b. Direct replicas of original features
c. Unmodified original features
d. Random noise components
Answer: a. Linear combinations of original featurespca4ds.github+2
How is explained variance ratio for each component calculated in PCA?
a. Eigenvalue of component / sum of all eigenvalues
b. Mean value / total features
c. Covariance / total variance
d. Feature count / sample count
Answer: a. Eigenvalue of component / sum of all eigenvaluespca4ds.github+1
PCA is most appropriate for which machine learning scenario?
a. High-dimensional input data with correlated features
b. Noisy datasets with binary features
c. Deep learning with unstructured text
d. Tabular data with independent features
Answer: a. High-dimensional input data with correlated featuresknowledge.dataiku+2
What preprocessing step can improve the interpretability of PCA results?
a. Standardization or normalization of features
b. Randomizing feature order
c. Removing all outliers
d. Duplicating data
Answer: a. Standardization or normalization of featuresappliedaicourse+1
Which of the following is NOT a fundamental limitation of PCA?
a. Assumes linearity between variables
b. Sensitive to scale of features
c. Requires all data to be labeled
d. Prone to information loss if too few components
Answer: c. Requires all data to be labeledpmc.ncbi.nlm.nih+2

Blog

Systemethic Technologies

Leave a Reply Cancel reply

Useful links

Other Terms & Policies

Contact Info