Context The SATO data set used is real life data collected from a major wireless telecom operator in South Asia. Content 1. **Aggregate of Total Revenue:** The overall monthly revenue earned in Rupees by the carrier in the months August & September 2015. 2. **Aggregate of SMS Revenue:** The revenue earned through the SMS service used by the subscriber. 3. **Aggregate of Data Revenue:** The revenue earned through the Data service used by the subscriber. 4. **Aggregate of Off Net Revenue:** The revenue earned by the calls etc. made to the off-network (not the same network as the subscriber) customers by the carrier’s present subscriber. 5. **Aggregate of On Net Revenue:** The revenue earned by the calls etc. made to the on-network (on the same network as the subscriber) customers by the carrier’s present subscriber. 6. **Network Age:** The time passed since the subscriber started using the services of the carrier. 7. **User Type:** This detail helps in knowing if the user is subscribed to a 2G or 3G service. 8. **Aggregate of Complaint Count:** The number of complaints made by the subscribers. 9. **Favorite Other Network:** This information can certainly have a huge impact on churn ratio as it gives the information about which other network or operator the subscribers makes the most of the calls to and thus might influence the customer to move to that network to save money. 10. **Aggregate of Data Volume:** The volume of the data service used by the subscriber. MCS: Multiple Classifier System to Predict the Churners in the Telecom Industry. Available from: https://www.researchgate.net/publication/320331663_MCS_Multiple_Classifier_System_to_Predict_the_Churners_in_the_Telecom_Industry [accessed Oct 12 2017]. Acknowledgements We wouldn't be here without the help of others. MCS: Multiple Classifier System to Predict the Churners in the Telecom Industry. Available from: https://www.researchgate.net/publication/320331663_MCS_Multiple_Classifier_System_to_Predict_the_Churners_in_the_Telecom_Industry [accessed Oct 12 2017]. Inspiration Why is the importance of a balanced dataset in churn prediction?