FinTech Innovations: Cloud-Enabled Financial Services for Scalable Digital Banking

Over the past few years, Financial Technology (FinTech) has revolutionized the financial services sector by improving the efficiency, automation, and scalability of loan approval systems. The financial sector moves increasingly towards automated decision-making, robust and scalable loan approval prediction systems become imperative. This work aims to propose and assess a cloud loan approval prediction system based on a Feedforward Neural Network (FNN) that enhances the prediction accuracy and scalability. Comprehensive data preprocessing techniques are integrated into the proposed methodology, involving missing value handling through imputation, scaling and normalization, and label encoding for categorical variables. This makes the data available for modelling, reducing bias and improving model accuracy. The FNN model is hosted and trained on Amazon Web Services (AWS) cloud infrastructure to allow the system to scale efficiently and process big data effectively while maintaining computational performance. The model attained exceptional performance, with precision, recall, accuracy, and F1-score of 0.9996, 0.9997, 0.9996, and 0.9996, reflecting its high accuracy in both loan approval and rejection predictions. The results strength is giving reliable real-time loan predictions, while the cloud infrastructure provides low latency and guarantees scalability for big financial applications. This research shows how cloud-based machine learning (ML) can make a considerable difference in loan approval processes by boosting automation and efficiency, providing a reliable solution for changing financial scenarios.