Published Paper Details:

Customers are the foundation for any business benefit and that is why firms become conscious of the significance of acquiring satisfaction of customers. Customer churn is one of the major problems and it is regarded as one of the most essential concerns among companies because of increasing among firms, increased significance of marketing policies and customers awareness in present years. Organizations must develop different policies to solve the churn issues depending on the services they offer. Customer churn practice is essential in competitive and rapidly developing in telecom sector. The process of changing from one service provider to another telecom service provider occurs due to good services or rates or due to benefits to the customers which the competitor firm provides customers when signing up. Due to the greater cost related with acquiring new customers the prediction of customer churn has developed as an indispensable part of planning process and strategic decision making in telecom sector. The main aim of the study is to explore the customer churn prediction in telecom using in big machine learning data platform. Machine learning techniques have been used for estimating the customer probability to churn. This study makes use of logistic regression and KNN with big data for predicting consumer churn in the telecom sector. Logistic regression has been used widely to estimate the probability of churn as a function of variables set or features of customers. Similarly, for churn K-Nearest Neighbour is used to examine if a customer churns or not based on their feature's proximity to customers in every class. This study uses Kaggle website for dataset in predicting and analysing churn