Decision Trees: A Comprehensive Guide for Machine Learning Students

Question 1

Which of the following is NOT a criterion used to select the optimal splitting attribute in a Decision Tree?

Accepted Answer

Classification Accuracy

Answer

Information Gain

Answer

Gini Impurity

Answer

Entropy

Question 2

What is the term for the process of removing unnecessary branches from a Decision Tree?

Accepted Answer

Pruning

Answer

Splitting

Answer

Merging

Answer

Overfitting

Question 3

What is the term for the maximum depth of a Decision Tree?

Accepted Answer

Height

Answer

Branching Factor

Answer

Width

Answer

Leaf Node

Question 4

What is the term for the node in a Decision Tree where a class label is assigned?

Accepted Answer

Leaf Node

Answer

Branch Node

Answer

Root Node

Answer

Terminal Node

Question 5

What is the process of splitting a node in a Decision Tree called?

Accepted Answer

Node expansion: Creating child nodes from a parent node based on a chosen splitting criterion

Answer

Partitioning: Dividing the data into subsets based on the values of the chosen attribute

Answer

Branching: Creating multiple paths from a parent node to child nodes representing different outcomes

Answer

Model construction: The process of building a Decision Tree from the root node to the leaf nodes

Question 6

Which splitting criterion evaluates the impurity of a node by calculating the entropy?

Accepted Answer

Information Gain: Measures the reduction in entropy after splitting a node based on an attribute

Answer

Gini Impurity: Calculates the probability of misclassification at a node based on class distribution

Answer

Chi-Square Test: Assesses the independence between an attribute and the target variable

Answer

F-Test: Compares the variance between groups to determine the significance of an attribute

Question 7

What is the purpose of pruning in Decision Trees?

Accepted Answer

To reduce overfitting and improve generalization performance by removing unnecessary branches

Answer

To increase the size and complexity of the tree to capture more complex patterns

Answer

To enhance the accuracy of the predictions by adding more decision nodes to the tree

Answer

To simplify the interpretation of the tree by removing redundant or irrelevant branches

Question 8

What is the name of the popular open-source Python library for implementing Decision Trees?

Accepted Answer

Scikit-learn: A comprehensive machine learning library that includes various Decision Tree algorithms and utilities

Answer

TensorFlow: Primarily used for deep learning, but can also be used to train Decision Trees through its Keras interface

Answer

NumPy: A library for numerical operations and data manipulation, not specifically designed for Decision Trees

Answer

Pandas: A library for data manipulation and analysis, not specifically designed for Decision Trees

Question 9

What is the process of removing redundant branches from a Decision Tree called?

Accepted Answer

Pruning

Answer

Splitting

Answer

Node reduction

Answer

Branching

Question 10

In a Decision Tree, the node that represents the final outcome is called the:

Accepted Answer

Leaf Node

Answer

Root Node

Answer

Internal Node

Answer

Decision Node

Question 11

A Decision Tree is said to be "overfitted" when:

Accepted Answer

It has too many branches

Answer

It performs poorly on unseen data

Answer

It has too few branches

Answer

It cannot handle missing values

Question 12

The process of recursively splitting the data to construct a Decision Tree is known as:

Accepted Answer

Tree Induction

Answer

Tree Pruning

Answer

Tree Evaluation

Answer

Tree Optimization

Question 13

What is the maximum depth of a Decision Tree with L levels?

Accepted Answer

L

Answer

L/2

Answer

2L

Answer

L^2

Question 14

Which of the following is NOT a limitation of Decision Trees?

Accepted Answer

Can handle large datasets efficiently

Answer

Prone to overfitting

Answer

Can suffer from high variance

Answer

Sensitive to missing values

Question 15

Which of the following is not a structural component of a Decision Tree?

Accepted Answer

Splitter nodes

Answer

Root node

Answer

Leaf nodes

Answer

Branch nodes

Question 16

In a Decision Tree, the attribute that leads to the highest information gain for data splitting is selected as the:

Accepted Answer

Internal node

Answer

Root node

Answer

Leaf node

Answer

Terminal node

Question 17

Which ensemble technique can be effectively combined with Decision Trees to improve their performance?

Accepted Answer

Random Forests

Answer

Support Vector Machines

Answer

K-Nearest Neighbors

Answer

Logistic Regression

Question 18

What is a major limitation associated with Decision Trees?

Accepted Answer

Susceptibility to overfitting

Answer

Complex and challenging to interpret

Answer

Computationally intensive during training

Answer

Cannot handle nonlinear relationships

Question 19

How does visualizing Decision Trees significantly benefit model comprehension?

Accepted Answer

Enhanced model comprehension

Answer

Reduced computational complexity

Answer

Accelerated deployment

Question 20

In Decision Tree construction, which impurity measure is NOT commonly used?

Accepted Answer

Mean squared error

Answer

Gini impurity

Answer

Entropy

Answer

Information gain

Question 21

When selecting a split attribute in Decision Tree construction, which of the following is a critical factor?

Accepted Answer

Information gain

Answer

Number of data instances in each branch

Answer

Depth of the tree

Answer

Mean absolute error of the split

Question 22

What is a key advantage of using Decision Trees for classification tasks?

Accepted Answer

Simplicity and interpretability of the model

Answer

Ability to represent non-linear decision boundaries

Answer

Robustness to outliers

Answer

Efficiency in handling high-dimensional datasets

Question 23

Which of the following is NOT a potential disadvantage associated with Decision Trees?

Accepted Answer

Robustness to missing values

Answer

Proneness to overfitting

Answer

Susceptibility to high variance

Answer

Difficulty in handling categorical features

Question 24

The process of dividing a node into child nodes based on the values of a selected feature is referred to as:

Accepted Answer

Splitting

Answer

Branching

Answer

Pruning

Answer

Learning

Question 25

The process of removing unnecessary branches from a Decision Tree to enhance its generalization ability is called:

Accepted Answer

Pruning

Answer

Splitting

Answer

Training

Answer

Validation

Question 26

Which of the following is a widely used method for selecting the best split point in a Decision Tree?

Accepted Answer

Greedy search

Answer

Random search

Answer

Exhaustive search

Answer

Brute-force search

Question 27

Which of the following is a potential limitation of using Decision Trees?

Accepted Answer

Prone to overfitting

Answer

Susceptible to underfitting

Answer

High computational cost during training

Answer

Inability to capture non-linear relationships

Question 28

Ensemble methods like Random Forests utilize multiple Decision Trees for improved accuracy. Which technique is employed to combine the predictions of these individual trees?

Accepted Answer

Bagging

Answer

Boosting

Answer

Stacking

Answer

Voting

Question 29

A Decision Tree with a maximum depth of 5 signifies that the tree cannot have more than:

Accepted Answer

5 levels

Answer

4 levels

Answer

6 levels

Answer

7 levels

Question 30

Which of the following is a popular open-source library for implementing Decision Trees in Python?

Accepted Answer

scikit-learn

Answer

TensorFlow

Answer

PyTorch

Answer

Keras

Question 31

How does setting a maximum depth for a tree act as a regularization technique in Decision Trees?

Accepted Answer

Limits excessive branching, reducing tree size and overfitting.

Answer

Enhances overfitting by allowing unrestricted tree growth.

Answer

Has no impact on overfitting in Decision Trees.

Question 32

How does the depth of a decision tree impact its accuracy? What method is used to find the best tree depth for a given dataset?

Accepted Answer

The optimal tree depth is found through techniques like cross-validation and using metrics like accuracy or F1 score.

Answer

Deeper trees always lead to higher accuracy.

Answer

Shorter trees are always the best choice for accuracy.

Answer

The number of features in the dataset determines the optimal tree depth.

Question 33

Which feature importance measure assesses the change in Gini impurity brought about by each feature in a Decision Tree?

Accepted Answer

Gini Importance

Answer

Information Gain

Answer

Gain Ratio

Answer

ReliefF

Question 34

A decision tree is primarily composed of nodes, branches, and:

Accepted Answer

Leaves

Answer

Vertices

Answer

Contours

Answer

Edges

Question 35

When growing a decision tree, the data is recursively partitioned into smaller subsets based on:

Accepted Answer

Feature values

Answer

Data points

Answer

Class labels

Question 36

Pruning a decision tree involves:

Accepted Answer

Removing redundant branches or nodes to prevent overfitting.

Answer

Expanding the tree to incorporate more data.

Answer

Altering the splitting criterion.

Question 37

A decision tree is classified as a:

Accepted Answer

Supervised learning algorithm

Answer

Generative learning algorithm

Answer

Unsupervised learning algorithm

Answer

Reinforcement learning algorithm

Question 38

The minimum sample size required at a leaf node for it to be deemed pure is termed:

Accepted Answer

Minimum Leaf Size

Answer

Minimum Node Size

Answer

Maximum Node Size

Answer

Maximum Leaf Size

Question 39

To evaluate a decision tree's performance on a regression task, an appropriate metric is:

Accepted Answer

Mean Absolute Error

Answer

Log Loss

Answer

F1-Score

Answer

Accuracy

Question 40

Random Forests utilize multiple:

Accepted Answer

Decision Trees

Answer

Neural Networks

Answer

Linear Regressions

Answer

Support Vector Machines

Question 41

Gradient Boosting Machines (GBMs) are an ensemble method that constructs:

Accepted Answer

Sequential Decision Trees

Answer

Independent Decision Trees

Answer

Nested Decision Trees

Answer

Parallel Decision Trees

Question 42

Which of the following is a defining characteristic of Decision Trees, a non-parametric machine learning algorithm?

Accepted Answer

Non-parametric

Answer

Supervised

Answer

Linear

Answer

Probabilistic

Question 43

Which algorithm is commonly employed for pruning Decision Trees?

Accepted Answer

Cost-Complexity Pruning

Answer

Forward Selection

Answer

Backward Elimination

Answer

Recursive Feature Elimination

Question 44

What is a key advantage of using Decision Trees for Machine Learning tasks?

Accepted Answer

Interpretability

Answer

Ability to handle vast datasets without compromising efficiency.

Answer

Minimal computational cost, requiring significantly fewer resources compared to other algorithms.

Answer

Exceptional accuracy, consistently outperforming other algorithms.

Question 45

Provide an example of a practical application where Decision Trees are commonly utilized in Machine Learning.

Accepted Answer

Fraud detection

Answer

Time series forecasting

Answer

Natural language processing

Answer

Image recognition

Question 46

Identify a potential limitation associated with Decision Trees.

Accepted Answer

Susceptibility to overfitting, which can compromise model performance on unseen data.

Answer

Inability to handle missing values, limiting their applicability to certain datasets.

Answer

Requirement for an extensive amount of training data, making them impractical for smaller datasets.

Question 47

Which of the following is **not** an advantage of Decision Trees?

Accepted Answer

High computational cost

Answer

Non-parametric nature

Answer

Robustness to noisy data

Answer

Straightforward to interpret

Question 48

What criterion is used in ID3 (Iterative Dichotomiser 3) to select the feature for splitting a node?

Accepted Answer

Information Gain

Answer

Entropy

Answer

Chi-Square Test

Answer

Gini Impurity

Question 49

What is the main distinction between Decision Trees constructed using pre-pruning and post-pruning techniques?

Accepted Answer

Pre-pruning builds the tree considering only current splits, while post-pruning builds the complete tree and then prunes.

Answer

Pre-pruned trees are always smaller than post-pruned trees.

Question 50

What is the purpose of the "root node" in a Decision Tree?

Accepted Answer

Represents the entire dataset and initiates the decision-making process

Answer

Represents the most frequent class label

Answer

Represents a final decision or prediction

Answer

Represents a split on a particular feature

Question 51

Which of the following is not a common tree traversal method?

Accepted Answer

Binary search

Answer

Pre-order

Answer

Depth-first

Answer

Breadth-first