Sentiment Analysis Based Telecom Churn Prediction

Abstract

Predicting customer churn is a big challenge and a survival basic for Telecom operators. In a large and competitive market like India, it is very essential to gather real-time customer feedback as a health indicator. Social networks have evolved as a rich source of real-time sentiments and opinions of the general public. In this research, tweets for the Twitter handle of 5 major telecom brands in India: Aircel, Bharti Airtel, Idea Cellular, Reliance Jio and Vodafone India were extracted for six months to develop a prediction model for telecom subscriber churn prediction using the sentiment score. Naïve Bayes classifier implementation and TextBlob library of Python were used to assign polarities to user sentiments. Customer satisfaction represented by the overall monthly sentiment score has been used to predict customer churn. The predictions made by the model were validated using IBM SPSS and were within the acceptable limits. The results of the sentiment analysis based prediction model can be of great use for telecom operators to take timely actions for improving the future customer experience and avoiding customer churn.

Keywords: Customer Churn, Naive Bayes, Opinion Mining, Sentiment Analysis, Social Network.

Cite this Article: Sandeep Ranjan, Sumesh Sood. Sentiment analysis based telecom churn prediction. Journal of Web Engineering & Technology. 2020; 7 (1): 6–12.