Building Scalable Fintech Platforms: Designing Secure and High Performance Mutual Fund and Loan Management Systems

Abstract

Designing scalable fintech platforms for mutual fund and loan management ensures robust performance, security, and flexibility to handle increasing transactions and user demands, making them crucial for efficient asset management and loan processing. However, a key drawback is the complexity of maintaining security standards, as scaling up often requires additional layers of protection, which lead to higher costs and implementation challenges. To overcome this problem, in this manuscript Building Scalable Fintech Platforms: Designing Secure and High-Performance Mutual Fund and Loan Management Systems (BSFP-SHML-PGCN) is proposed.  The major objective of the proposed method is to secure financial fund and loan management.Initially, the input data are collected from Mutual Funds India–Detailed Dataset. The data are pre-processed using Multiparticle Kalman Filter (MKF), which are used missing values and clean input data. After that, the data are fed into Hyperbolic Cosine Transform (HCT)for extract relevant features such as scheme name, expense ratio, rating, and category. The extracted features are provided toProgressive Graph Convolutional Networks (PGCN) to classify theuncertainty in mutual fund returns risk level as Low Risk, Low to Moderate, Moderate, Moderately High, High, and Very High.The proposed technique is implemented in Python, and the efficacy of the BSFP-SHML-PGCN technique was assessed using various performance measures, including accuracy, precision, sensitivity, and specificity. The performance of the BSFP-SHML-PGCN method achieved99.05% higher accuracy, 99.01% higher precision, 98.95% higher sensitivity, and 96.50% higher specificity when analysed through existing techniques such as FinTech enablers, use cases, and role of future internet of things (FTE-FIoT-DL), Investigating the components of fntech ecosystem for distributed energy investments and an integrated quantum spherical decision support system (ICFE-DEI-SWARA) and Forecasting the returns of the US real estate investment trust market: evidence from the group model of data handling neural network (USRE-ITME-ML),  respectively.