Predicting Credit Risk Defaulting in Peer - To - Peer(P2P)Lending with Deep Learning

Victor Bwalya; Nancy Namonje; Muwanei Sinyinda (PhD)

doi:10.36948/ijfmr.2026.v08i05.85181

Predicting Credit Risk Defaulting in Peer - To - Peer(P2P)Lending with Deep Learning

Author(s)	Victor Bwalya, Nancy Namonje, Muwanei Sinyinda (PhD)
Country	Zambia
Abstract	Peer-to-peer (P2P) lending has gained popularity as a method of lending money in recent years. It has emerged as a disruptive force in the financial industry, connecting borrowers directly with lenders. However, accurate credit risk assessment is crucial to P2P lending's viability and success. The complex and dynamic relationships included in P2P lending data may be difficult for traditional credit risk defaulting models to capture. In the context of peer-to-peer lending, this study suggests a novel strategy to improve credit risk default prediction by utilizing deep learning techniques with feature engineering. Due to the burdensome application processes and excessive interest rates, many customers are unable to get personal loans from banks. Because of this, potential borrowers are still looking for other ways of getting a loan, and that’s where P2P lending comes in. The primary focus of this research was to develop a deep learning model specifically tailored for predicting credit risk defaulting in P2P lending by making use of feature engineering techniques, assess the model's ability to capture complex patterns, and compare its performance against traditional credit scoring models. The research methodology involved the collection of historical loan data from a representative of P2P lending platform, comprehensive data preprocessing (feature engineering techniques), and the implementation of deep learning model - CNNs (Convolution Neural Networks). When training the model, predictive accuracy and risk sensitivity were optimized. Performance criteria like accuracy, precision, recall, F1-score, and AUC-ROC were used to assess the trained proposed model. 99.93% had the highest accuracy rate, which was superior compared to the other models. AUC was 90.96%, recall was at 72%, and finally precision was at 88%. The findings indicated that the suggested CNN-based prediction model performed better when incorporated with feature engineering techniques, to predict credit risk defaulting in peer-to-peer lending. This will also help reduce human bias in credit decision-making, leading to fairer outcomes for borrowers from diverse backgrounds.
Keywords	Peer-to-Peer (P2P) lending, Deep Learning, Convolutional Neural Network (CNN), Credit Risk Scoring, Model evaluations, Feature Engineering Techniques
Field	Computer
Published In	Volume 8, Issue 5, September-October 2026
Published On	2026-09-09
DOI	https://doi.org/10.36948/ijfmr.2026.v08i05.85181

View / Download PDF File

About IJFMR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9687-828-838	editor@ijfmr.com

International Journal For Multidisciplinary Research

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Bi-monthly Scholarly International Journal

Predicting Credit Risk Defaulting in Peer - To - Peer(P2P)Lending with Deep Learning

Share this