Decision Trees in Machine Learning

Play an AI-generated podcast conversation about this lesson

What is the main purpose of using entropy in classification trees?

To quantify missing information (correct)

To determine the speed of the classification

To simplify data representation

To increase the complexity of the data

What classification criterion is used in classification trees?

The shape of the objects only

Only the size of the objects

Only the color of the objects

Both shape and color of the objects (correct)

What does a good test in a classification tree indicate?

It is simple to answer

It leads to many conclusions

It requires more than two options

It carries much information about the class (correct)

What encoding is suggested for more frequent classes in information theory?

Using shorter encoding for frequent classes Signup and view all the answers

Why might one stop splitting nodes in a decision tree?

When further classification does not yield additional information Signup and view all the answers

What type of decision tree is used when Y is a nominal variable?

Classification tree Signup and view all the answers

What is one of the main reasons for using decision trees over other models?

They are interpretable and can explain predictions. Signup and view all the answers

In decision tree learning, what does the loss function ℓ signify?

The penalty for incorrect predictions. Signup and view all the answers

What is the goal in the second learning task when using decision trees?

To minimize the risk of predictions on unseen data. Signup and view all the answers

Why is learning decision trees considered NP-hard?

Finding the smallest tree consistent with data is complex. Signup and view all the answers

What does recursive partitioning in decision trees involve?

Dividing data subsets until all subsets are completely pure. Signup and view all the answers

What is the primary assumption made when learning decision trees from data?

Data contains examples of the output function. Signup and view all the answers

What characterizes a regression tree in decision tree learning?

Y is numerical. Signup and view all the answers

Which of the following is NOT a characteristic of decision trees?

They are always the most accurate predictive model. Signup and view all the answers

What is often used to find a suitable decision tree when the risk cannot be computed?

Heuristics and approximations. Signup and view all the answers

What is the primary purpose of a decision tree?

To represent a decision-making procedure Signup and view all the answers

In a decision tree, what do the branches represent?

The tests or questions guiding the decision Signup and view all the answers

What does the output attribute Y in a decision tree represent?

The target value assigned by the tree Signup and view all the answers

How does a decision tree handle a continuous input attribute?

It forms ranges or intervals instead of individual values Signup and view all the answers

Which of the following best describes the mapping function of a decision tree?

It maps input attributes to a unique output Signup and view all the answers

What kind of functions can be represented by decision trees if the input attributes are boolean?

All possible boolean functions Signup and view all the answers

Which of these examples does NOT represent a boolean function that can be depicted by a decision tree?

X1 + X2 (Arithmetic sum) Signup and view all the answers

What role do the input attributes X1, X2, …, Xn play in a decision tree?

They define the input space and tested conditions Signup and view all the answers

What does entropy measure in the context of a set of objects?

The amount of uncertainty regarding the class of an instance Signup and view all the answers

In the equation for class entropy CE(S), what does pi represent?

The proportion of elements in S of class ci Signup and view all the answers

What is the expected outcome when performing a question with high expected information gain?

Reduction of entropy Signup and view all the answers

Given a set with classes A, B, and C, which scenario would result in the highest entropy?

Equal distribution of A, B, and C Signup and view all the answers

What computation can be used to determine the information gain from a test in classification trees?

CE(S) - E[CE(Si)] Signup and view all the answers

In the context of class entropy computation, what would be the effect of a class distribution with a high number of instances in one class?

Decrease in entropy Signup and view all the answers

How is class entropy defined mathematically?

<ul> <li>∑ pi log2(pi)</li> </ul> Signup and view all the answers

What is indicated by high entropy in a dataset?

Many possible outcomes, all equally likely Signup and view all the answers

What is the relationship between entropy and information gain?

Information gain represents the reduction in entropy Signup and view all the answers

What is the value of entropy when a set contains 15 instances of class A and 1 instance of class B?

0.34 Signup and view all the answers