Subscriber Churn Analysis (Telecom)

The main objective of this case study is to analyze and overcome the customer churn in the telecom business. Hence we can predict early in advance that at what stage particular customer churns the company so as we can make a strategy against a particular individual from past data and overcome the preventing churn in the company.

To carry out the process, the first analysis is done to gain the important features of this project and visualize them. what are the proportions for the customer who has an international plan and churn same as we can analyze for a Customer Service call and derive the conclusion that the particular attributes are important for our modeling or not? Also, we check for multicollinearity amongst the predictors and it also helps to see the relationship between the attributes. Using this kind of technology we use for analyzing the data and feature importance. Logistic regression was used to find out the weights of the features.