• M.G.V.S.Sreeja
• B.sai Deepak
• Ch.Sai Jayanth
• D.Kalyan Kumar
• D.Chandrika

Customer churn prediction is gaining popularity in business, particularly in the telecommunications industry. Many models have presented various versions of the churn prediction models that are heavily based on data mining concepts and employ machine learning and metaheuristic algorithms. The goal of this project is to create the most important churn prediction model. The primary goal of churn in the telecom industry is to accurately estimate the customer survival and customer hazard functions in order to gain a complete understanding of churn over the customer tenure. Another goal is to identify customers on the verge of churn and predict when they will leave. This project focuses on analysing churn prediction techniques in order to identify churn behaviour and validate the reasons for customer churn. Customer churn prediction and survival analysis are important areas of research in business analytics. Churn prediction involves identifying customers who are at risk of discontinuing their relationship with a company, while survival analysis focuses on modeling the time to an event (such as churn) for a group of customers. Both of these tasks are essential for businesses that want to maintain customer loyalty and reduce customer turnover. Machine learning algorithms such as Cox Proportional Hazards Model, Random Forests, Logistic Regression, Gradient Boosting, and Neural Networks can be used to build models for churn prediction and survival analysis. The choice of evaluation metrics for these models depends on the specific goals and requirements of the analysis. Common metrics for churn prediction models include accuracy, precision, recall, F1 Score, and ROC Curve, while metrics for survival analysis models include Concordance Index (C-index), Log-Rank Test, Brier Score, AUC, and Integrated Brier Score. By leveraging these models and metrics, businesses can gain insights into customer behavior, develop targeted retention strategies, and ultimately improve their bottom line.

Churn, Survival ,Telecom, Random Forest

"Customer Survival And Attrition Prediction", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 4, page no.e149-e155, April-2023, Available :http://www.jetir.org/papers/JETIR2304422.pdf

"Customer Survival And Attrition Prediction", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 4, page no. ppe149-e155, April-2023, Available at : http://www.jetir.org/papers/JETIR2304422.pdf