Telecomm Churn Prediction Using Fundamental Classifiers To Identify Cumulative Probability

Churn is one of the most important and desired use-cases in Business. Churn rate is defined as the percentage of subscribers to a service that switches their subscription to another in a period frame. Organizations and other industries use this factor to measure the rate or amount at which customers quit product sites or Subscription services such as Amazon, Netflix, etc. Churn also occurs in Telecommunication Industry where customers switch their sim cards from one point to another. In this paper, We are going to predict foamin the Telecomm Sector, Where the Data consists of 226 Feature Samples. Considering the 3 Months and other call details such as Incoming calls, Outgoing calls,2g,3g data used. We will be predicting the churn that Occurred in the 4^th i. e last month from the dataset. This research aims to find the Predictive model for churn along with finding churn Probability, Important features, and Strong Predictors of churn from the data. We have Implemented Classifier algorithms such as Extra-trees, XG- boost, Random forest, decision tree, Logistic Regression for predicting the churn.