The churn rate, also known as the rate of attrition, is the percentage of subscribers to a service who discontinue their subscriptions to that service within a given time period. For a company to expand its clientele, its growth rate, as measured by the number of new customers, must exceed its churn rate. The rate is generally expressed as a percentage.
This is data analysis project for Tele2 Big Data Academy 2018. For this task I chose a telecom company customers churn data set and applied several methods for data cleaning, feature engineering and exploratory analysis. After understanding main trends and criteria leading to customer churn, I tried some popular machine learning algorithms to predict potential churners on unseen testing data. For this task I used well-known Machine Learning algorithms (Logistic Regression, SVM, Tree-based models, KNN, Neural Networks). In order to have full control of prediction process, I implemented feed-forward Neural Network model as well Dropout and Adam optimizer algorithms from scratch only in base R.
