Training a binary classification model using the XGBoost algorithm

Binary classification models are used to solve the problem of predicting one class of two possible classes – for example, predicting whether it will rain or not. The goal is to learn about past data points and figure out which one of the target buckets a particular data point will fall into. The typical use cases of a binary classification model are as follows:

• Predicting whether a patient suffers from a disease
• Predicting whether a customer will churn or not
• Predicting behavior – for example, whether a customer will file an appeal or not

In the next few sections, we will go through the following steps to achieve our goal of creating a binary classification model to be used to run inference queries:

1. Defining the business problem
2. Uploading and analyzing data
3. Creating the model
4. Running prediction queries against the model

Establishing the business problem...