In this section, we are going to discuss how to use an ANN model to predict the customers at risk of leaving or customers who are highly likely to churn. By the end of this section, we will have built a customer churn prediction model using the ANN model. We will be mainly using the dplyr, ggplot2, and keras libraries to analyze, visualize, and build machine learning models. For those readers who would like to use Python, instead of R, for this exercise, see the previous section.
For this exercise, we will be using one of the publicly available datasets from the IBM Watson Analytics community, which can be found at this link: https://www.ibm.com/communities/analytics/watson-analytics-blog/predictive-insights-in-the-telco-customer-churn-data-set/. You can follow this link and download the data, which is available in XLSX format, named WA_Fn-UseC_...
