Question

You are creating a binary classification model based on a DataFrame that contains the following data about loan applications: customer_id income savings debt approve 105671 75000 1750 26000 no 105672 120000 50750 27750 yes 105673 54000 37000 3500 yes You want to use the data in the income, savings, and debt fields as features to predict a label based on the approve field. Which of the following changes must you make to the label data before using it to train the model?

A. Convert the values in the approve column to numeric representations of the values "yes" and "no".B. Convert the values in the approve column to a vector.C. Rename the approve column to label.D. Create a new DataFrame with a single column that contains a vector of the income, savings, debt, and approve values.

Answer 1

A: Convert the values in the approve column to numeric representations. That is "Yes" = 0 and "No" = 1

Step-by-step explanation:

Binary classification problems seeks to classify an object into one of two classes. Usually each class has a label like spam or not spam, churn or not churn. The task is usually to predict that a new case belongs to one of the two classes using machine learning algorithms.

Since the task is to classify "yes" or "no" these labels are assigned numeric values of 0 and 1.

Popular machine learning algorithms for binary classifications include:

Decision trees, Logistic regression, support vector machine, Naive Bayes and K-Nearest Neighbour.