Feature Selection and Extraction: Master Dimensionality Reduction and Feature Engineering

Question 1

Which of the following is NOT a primary objective of feature selection techniques?

Accepted Answer

Increase feature cardinality

Answer

Identify crucial features

Answer

Enhance interpretability

Answer

Reduce dimensionality

Question 2

What is the main advantage of using supervised feature selection techniques over unsupervised ones?

Accepted Answer

They incorporate information about the target variable during feature selection.

Answer

They can be applied to both numerical and categorical features.

Answer

They are computationally more efficient.

Question 3

What is the main difference between principal component analysis (PCA) and singular value decomposition (SVD)?

Accepted Answer

SVD can handle both numerical and categorical features, while PCA is limited to numerical features.

Answer

PCA is a linear transformation, while SVD is a non-linear transformation.

Answer

PCA is unsupervised, while SVD is supervised.

Question 4

Which of the following is a potential disadvantage of using embedded methods for feature selection?

Accepted Answer

They can be computationally expensive, especially for large datasets.

Answer

They are prone to overfitting.

Answer

They cannot be used with non-linear models.

Question 5

Which of the following is a common metric used to assess the performance of a dimensionality reduction technique?

Accepted Answer

Cumulative explained variance

Answer

R2 score

Answer

Mean squared error

Answer

Log loss

Question 6

Which of the following is NOT a criterion typically used for feature selection?

Accepted Answer

Cost of data collection

Answer

Redundancy

Answer

Relevance

Answer

Informativeness

Question 7

Which feature selection method employs a greedy approach to iteratively select features?

Accepted Answer

Sequential Forward Selection

Answer

Principal Component Analysis

Answer

Tree-Based Methods

Answer

L1 Regularization

Question 8

What is a key benefit of utilizing dimensionality reduction techniques in machine learning?

Accepted Answer

Enhanced model interpretability, reduced overfitting, and improved computational efficiency.

Answer

Increased model complexity

Answer

Always improved predictive performance

Question 9

Which of the following is a widely used technique for feature extraction?

Accepted Answer

Principal Component Analysis

Answer

Support Vector Machines

Answer

Linear Regression

Answer

Decision Trees

Question 10

What is the primary goal of filter methods in feature selection?

Accepted Answer

To evaluate and rank features based on individual characteristics without considering interactions.

Answer

To identify the most relevant features for a specific task considering interactions.

Answer

To reduce the computational cost of model training.

Question 11

Which of the following is a potential limitation of wrapper methods in feature selection?

Accepted Answer

High computational cost, potential for overfitting, and sensitivity to noisy data.

Answer

Limited ability to handle large datasets.

Answer

Difficulty in interpreting the selected features.

Question 12

When evaluating the performance of feature selection methods, which factor is most crucial to consider?

Accepted Answer

The trade-off between model accuracy, feature reduction, and computational efficiency.

Answer

Computational efficiency

Answer

Interpretability of the selected features

Question 13

What is the primary advantage of incorporating domain knowledge in feature selection?

Accepted Answer

Improved feature relevance, reduced data exploration time, and better alignment with the problem domain.

Answer

Enhanced predictive performance

Answer

Increased computational efficiency

Question 14

Which of the following statements about feature selection is INCORRECT?

Accepted Answer

Feature selection always improves model performance.

Answer

Feature selection can reduce the risk of overfitting.

Answer

Feature selection can simplify models and make them easier to interpret.

Answer

Feature selection can reduce training time.

Question 15

Which of the following is a common approach for evaluating the effectiveness of different feature subsets?

Accepted Answer

Cross-validation

Answer

Regression analysis

Answer

Principal component analysis

Answer

Linear discriminant analysis

Question 16

Which of the following feature extraction techniques involves reducing the dimensionality of the original feature space?

Accepted Answer

Principal component analysis

Answer

Feature hashing

Answer

Feature scaling

Answer

One-hot encoding

Question 17

The process of identifying and selecting a subset of features that contribute most to the predictive performance of a machine learning model is referred to as:

Accepted Answer

Feature selection

Answer

Data mining

Answer

Model training

Answer

Feature transformation

Question 18

Which of the following techniques can be utilized to combine similar features into a single, representative feature?

Accepted Answer

Feature agglomeration

Answer

Embedded method

Answer

Filter method

Answer

Wrapper method

Question 19

A feature selection method that evaluates individual features based on their contribution to the prediction task is known as a:

Accepted Answer

Filter method

Answer

Wrapper method

Answer

Embedded method

Question 20

Which feature extraction technique involves transforming the original feature space into a new, lower-dimensional space while preserving important information?

Accepted Answer

Projection

Answer

Selection

Answer

Discretization

Answer

Aggregation

Question 21

The technique of converting categorical features into a binary representation, where each category is represented by a unique binary vector, is called:

Accepted Answer

One-hot encoding

Answer

Hashing

Answer

Binarization

Answer

Label encoding

Question 22

Which of the following factors should NOT be considered when selecting features for a machine learning model?

Accepted Answer

Computational resources available

Answer

Data quality

Answer

Model interpretability

Answer

Predictive power of features

Question 23

Which of the following is NOT a filter method for feature selection?

Accepted Answer

Wrapper method

Answer

Information Gain

Answer

Chi-square test

Question 24

What is the primary purpose of dimensionality reduction in feature selection?

Accepted Answer

To reduce the number of features while retaining the most important information.

Answer

To improve the interpretability of a machine learning model.

Answer

To increase the accuracy of a machine learning model.

Question 25

What is the fundamental difference between feature selection and feature extraction?

Accepted Answer

Feature selection removes irrelevant features, while feature extraction creates new ones.

Answer

Feature selection is more computationally expensive than feature extraction.

Answer

Feature selection is a supervised technique, while feature extraction is an unsupervised technique.

Question 26

What is the technical term for the challenge faced when working with high-dimensional datasets?

Accepted Answer

Curse of dimensionality

Answer

Overfitting

Question 27

Which feature selection method ranks features based solely on their individual predictive power?

Accepted Answer

Filter method

Answer

Wrapper method

Answer

Embedded method

Answer

Hybrid method

Question 28

What is the primary goal of feature extraction in machine learning?

Accepted Answer

To derive a new set of representative features from the original data

Answer

To eliminate redundant or irrelevant features

Answer

To enhance the computational efficiency of models

Question 29

Which feature extraction technique involves utilizing linear combinations of the original features?

Accepted Answer

Principal Component Analysis (PCA)

Answer

t-SNE

Answer

Autoencoder

Answer

Linear Discriminant Analysis (LDA)

Question 30

Which evaluation metric is commonly used to assess the performance of feature selection methods?

Accepted Answer

F-score

Answer

Recall

Answer

Precision

Answer

Accuracy

Question 31

What is the primary role of cross-validation in feature selection?

Accepted Answer

To prevent overfitting and ensure the selected features generalize well

Answer

To accelerate the feature selection process

Answer

To determine the ideal number of features to be selected

Question 32

Explain the fundamental difference between forward and backward feature selection algorithms.

Accepted Answer

Forward selection starts with an empty set, iteratively adding features, while backward selection begins with the full set, iteratively removing features

Answer

Forward selection is greedy while backward selection is exhaustive

Question 33

What is the primary aim of feature standardization?

Accepted Answer

To transform features such that they have a mean of 0 and standard deviation of 1

Answer

To remove outliers from data

Answer

To reduce the dimensionality of data

Question 34

Which method of feature selection considers the interaction between features and the target variable?

Accepted Answer

Wrapper method

Answer

Filter method

Answer

Embedded method

Question 35

What is a key advantage of utilizing mutual information for feature selection?

Accepted Answer

It can capture non-linear relationships between features and the target variable

Answer

It is insensitive to outliers

Answer

It is computationally fast

Question 36

Which technique is primarily employed for dimensionality reduction in feature extraction?

Accepted Answer

Principal Component Analysis (PCA)

Answer

Support Vector Machine

Answer

K-Means Clustering

Question 37

What is the primary objective of feature scaling?

Accepted Answer

To standardize the range of feature values for enhanced comparability

Answer

To reduce the dimensionality of the dataset

Answer

To eliminate outliers from the dataset

Question 38

What potential drawback is associated with embedded feature selection methods?

Accepted Answer

They can exhibit bias towards the specific machine learning algorithm used

Answer

They are computationally intensive

Answer

They are restricted in the number of features they can select

Question 39

What is the crucial role of a domain expert in the feature selection and extraction process?

Accepted Answer

To provide domain-specific knowledge for identifying relevant and meaningful features

Answer

To develop and implement feature selection algorithms

Answer

To evaluate the performance of different feature selection techniques

Question 40

What is the primary function of a filter method in feature selection?

Accepted Answer

To efficiently identify features with high statistical relevance to the target variable

Answer

To reduce the dimensionality of the dataset

Answer

To select features that are directly pertinent to a specific machine learning algorithm

Question 41

Which of the following is an example of a heuristic feature selection method?

Accepted Answer

ReliefF

Answer

Linear Regression

Answer

Support Vector Machine

Answer

Decision Tree

Question 42

What is the fundamental goal of feature extraction in the context of machine learning?

Accepted Answer

To create new features that capture the underlying structure and representation of the data

Answer

To reduce the dimensionality of the data

Answer

To select the most important features for a specific task

Question 43

Which of the following is NOT a benefit of feature selection?

Accepted Answer

Increased complexity of the model

Answer

Enhanced interpretability

Answer

Improved model performance

Answer

Reduced training time

Question 44

Which feature selection technique utilizes statistical tests to identify features with strong relationships to the target variable?

Accepted Answer

Univariate Feature Selection

Answer

Wrapper Methods

Answer

Dimensionality Reduction

Answer

Embedded Methods

Question 45

What is the main difference between feature selection and feature extraction?

Accepted Answer

Feature selection chooses existing features, while feature extraction creates new features.

Answer

Feature selection is unsupervised, while feature extraction is supervised.

Answer

Feature selection reduces dimensionality, while feature extraction increases it.

Question 46

A researcher is analyzing a dataset with 100 features. Which technique would be most appropriate if the goal is to reduce the number of features to 10 while preserving as much variance as possible?

Accepted Answer

Principal Component Analysis (PCA)

Answer

Recursive Feature Elimination (RFE)

Answer

Forward Feature Selection

Question 47

You are building a model to predict customer churn. Which of the following is likely to be a more effective feature engineering strategy?

Accepted Answer

Creating new features that capture customer engagement metrics, such as average transaction value or number of active subscriptions.

Answer

Removing all features related to customer demographics.