Telecom Customer Churn Analysis Using Machine Learning

Customer churn is a major problem that many telecommunication service providers face in real-time. They incur a lot of financial loss when their customers switch their service to their competitors. Many analysis show that predicting churn is one of the best way to retain their own customers. With the upcoming mass use of Big Data in the present scenario, many machine learning models have been proposed to solve this issue. Yet, the challenges like performance and accuracy remain. Through this paper, challenges faced have been addressed by processing a publicly available dataset to analyze the cause of churn. The PySpark was utilized to start a Spark session in which we normalize and apply machine learning models into the dataset and find the factor that most likely affects the churn to provide insights to the telecom services on what to give more focus on. The results of this research highlights the important factors that increase churn rate enabling the service providers to prevent customer loss.