Enhanced Customer Churn Prediction using a Resnet-LSTM Hybrid Approach

Customer churn prediction is a crucial aspect of business intelligence that enables companies to proactively retain customers and enhance operational efficiency. This research proposes a hybrid deep learning model integrating Residual Networks (ResNet) and Long Short-Term Memory (LSTM) to improve churn prediction accuracy. The ResNet component extracts high-level static features, while the LSTM captures temporal patterns in customer behavior. By leveraging both spatial and sequential dependencies in customer data, the model provides enhanced predictive capabilities compared to traditional machine learning methods. The study involves data collection, preprocessing, feature extraction, model training, evaluation, and optimization. The results demonstrate that the hybrid model significantly improves accuracy, recall, and F1-score, making it a powerful tool for businesses to identify at-risk customers and implement targeted retention strategies. The findings contribute to the advancement of AI-driven customer relationship management and churn mitigation techniques.