Supervised Learning Quiz

What are the two basic machine learning methods introduced in the chapter?

• Clustering and Neural Networks

• k-nearest neighbors and decision trees

• Linear Regression and SVM

• Logistic Regression and Random Forest

Answer explanation

The two basic machine learning methods introduced in the chapter are k-nearest neighbors and decision trees.

In supervised machine learning, what does the training data contain?

• Examples of observed pairs of input and output values

• Only output variables

• Only input variables

• Labels for the input variables

Answer explanation

Training data in supervised machine learning contains examples of observed pairs of input and output values, allowing the model to learn the relationship between them.

What is the term used to describe the process of predicting the output for new, unseen test data?

• Inference

• Extrapolation

• Interpolation

• Generalization

Answer explanation

The term used to describe the process of predicting the output for new, unseen test data is Generalization.

What is the primary distinction between numerical and categorical variables?

• Numerical variables have no natural ordering

• Categorical variables are always continuous

• Categorical variables have a natural ordering

• Numerical variables are always discrete

Answer explanation

The primary distinction between numerical and categorical variables is that numerical variables have no natural ordering.

What is the main goal of supervised machine learning?

• To reason about the connection between input and output variables

• To make accurate predictions for new data points

• To visualize the training data

• To identify outliers in the data

Answer explanation

The main goal of supervised machine learning is to make accurate predictions for new data points.

What is the primary reason for using the k-nearest neighbours method?

• It can handle high-dimensional data well

• It can make predictions based on similar data points

• It is a parametric method

• It is computationally efficient

Answer explanation

The primary reason for using the k-nearest neighbours method is that it can make predictions based on similar data points.

What is the main drawback of the 1-nearest neighbour method?

• It is not suitable for regression problems

• It is a parametric method

• It relies on only one data point for prediction

• It is too complex to implement

Answer explanation

The main drawback of the 1-nearest neighbour method is that it relies on only one data point for prediction.

In the example of predicting colours with k-NN, what is the output variable being predicted?

• Artist Name

• Speed

• Temperature

• Colour

Answer explanation

The output variable being predicted in the example of predicting colours with k-NN is 'Colour'.

What is the purpose of normalizing input data in the k-NN method?

• To make the data more visually appealing

• To improve the accuracy of predictions

• To increase the complexity of the model

• To decrease the training time

Answer explanation

Normalizing input data in the k-NN method improves the accuracy of predictions by ensuring all features have the same scale, preventing any one feature from dominating the distance calculation.

What is the main difference between a leaf node and an internal node in a decision tree?

• A leaf node is the starting point of the tree

• An internal node contains the final prediction value

• An internal node is the endpoint of the tree

• A leaf node contains rules for splitting the data

Answer explanation

A leaf node contains rules for splitting the data, while an internal node does not contain the final prediction value.