Modeling and customer churn prediction using deep learning

• Main Authors: Sundaram, Gayathri, Reddy, Venkateswar, Reddy, Thirupathi, Reddy, Rakesh
• Format: Conference Proceeding
• Language: English
• Subjects: Brand loyalty; Clustering; Corporate learning; Customers; Datasets; Deep learning; Marketing; Telecommunications
• ISSN: 0094-243X; 1551-7616
• DOI: 10.1063/5.0208736

For a business to succeed, customer loyalty is crucial. The retention of loyal customers is a primary goal for any business. The success of companies, which rely on recurring revenue, depends on keeping their customers happy. Costs to acquire new customers are high. Super costly. This makes retaining them essential, regardless of the size of your market. Attempting to out-acquire a rising churn rate is futile. The importance of knowing the meaning of the word "loyalty" becomes clear when a company’s success hinges on it. The recent progress made in representation learning provides an opportunity to streamline and generalize feature engineering for all kinds of different applications. This study uses a real-world telecom dataset to predict customer churn and suggests using boosting to improve existing models. In this research, a feature set optimized LSTM is proposed to predict telecom customer churn on Telecom dataset. To ensure maximum profit, we also offer some marketing strategies in line with the clustering results and simulate a simplified marketing activity.