Hyper Personalized Risk Assessment for Fintech Using GEN AI

DOI

10.2991/978-94-6239-616-6_55How to use a DOI?

Keywords

Credit Risk Prediction; Stacked Ensemble; SMOTE-ENN; Deep Learning; Logistic Regression; Random Forest; CNN; MLP; Class Imbalance; Financial Risk Modelling; SME Credit Scoring

Abstract

Traditional credit risk assessment models often underperform when applied to real-world data from medium and micro enterprises (MMEs), primarily due to data imbalance, noisy financial records, and limited historical credit information [7, 12]. In this study, we propose a deep learning–based framework for credit risk prediction that employs a stacked ensemble approach integrating Random Forest, Logistic Regression, and Convolutional Neural Network (CNN) as base learners, and a Multi-Layer Perceptron (MLP) is used as the meta-learner to merge and refine their predictions [1, 5]. The model adopts a hybrid resampling strategy, combining SMOTE with Edited Nearest Neighbours (SMOTE-ENN) to balance uneven class distributions and reduce bias in credit risk datasets [1]. The financial feature set is designed to reflect time-dependent trends and borrower behavioral patterns [5, 7]. The ensemble system is trained and validated on bench-mark datasets as well as real-world SME financial records, achieving improved sensitivity and precision in identifying high-risk borrowers [1, 5]. This robust and scalable solution strengthens credit screening, helping lenders and financial institutions reduce defaults and enhance decision-making for underserved enterprise segments [11, 12].

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - B. Vijayakumar
AU  - D. Prasanaa Venkatesh
AU  - V. Sailesh
AU  - A. Ariharan
PY  - 2026
DA  - 2026/03/31
TI  - Hyper Personalized Risk Assessment for Fintech Using GEN AI
BT  - Proceedings of the International Conference on Artificial Intelligence and Secure Data Analytics (ICAISDA 2025)
PB  - Atlantis Press
SP  - 732
EP  - 742
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6239-616-6_55
DO  - 10.2991/978-94-6239-616-6_55
ID  - Vijayakumar2026
ER  -