Customer Churn Prediction (CCP) is a crucial procedure for maintaining clients, according to a number of businesses. Customer retention is a significant difficulty in many business sectors. Churn analysis has been used for a long time to improve customer-company relationships and profitability. Since there are more providers of telecommunications services, CCP in the telecommunications industry has become an essential requirement. Large businesses face a significant problem with customer churn because it is much more important to retain existing customers than to acquire new ones.Businesses will be able to keep their current customers and acquire new customers by following a predictable customer model. The selection of customer attributes (feature selection) from the dataset for the model's construction is crucial to the model's efficiency. Effective CCP models have only recently started to be developed using Deep Learning (DL) and Machine Learning (ML) models.A novel feature selection strategy based predictive modeling for customer churn prediction in telecom industry using swish CNN (Convolutional Neural Network) is presented in this analysis.Finally, a modified version of the S-CNN (Swish Convolutional Neural Network) is used to distinguish between a CC (Churn Customer) and a normal customer. If there is a churn, the customer retention process looks at the network utilization history.This approach's performance is evaluated in terms of Precision, Accuracy, F1-score, and Sensitivity.