Prediction of loan default based on multi-model fusion

Li, Xingyun; Ergu, Daji; Zhang, Di; Qiu, Dafeng; Cai, Ying; Ma, Bo

doi:10.1016/j.procs.2022.01.094

Cited by 10 publications

(4 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[15] offered the under-sampling method in their resampling ensemble model called REMDD for imbalanced credit risk evaluation in P2P lending. In the work [16], the ADASYN (adaptive synthetic sampling approach) [17] was adopted for reducing the class imbalance problem. Meanwhile, ref.…”

Section: Literature Reviewmentioning

confidence: 99%

Enhancing Supervised Model Performance in Credit Risk Classification Using Sampling Strategies and Feature Ranking

Wattanakitrungroj,

Wijitkajee,

Jaiyen

et al. 2024

BDCC

View full text Add to dashboard Cite

For the financial health of lenders and institutions, one important risk assessment called credit risk is about correctly deciding whether or not a borrower will fail to repay a loan. It not only helps in the approval or denial of loan applications but also aids in managing the non-performing loan (NPL) trend. In this study, a dataset provided by the LendingClub company based in San Francisco, CA, USA, from 2007 to 2020 consisting of 2,925,492 records and 141 attributes was experimented with. The loan status was categorized as “Good” or “Risk”. To yield highly effective results of credit risk prediction, experiments on credit risk prediction were performed using three widely adopted supervised machine learning techniques: logistic regression, random forest, and gradient boosting. In addition, to solve the imbalanced data problem, three sampling algorithms, including under-sampling, over-sampling, and combined sampling, were employed. The results show that the gradient boosting technique achieves nearly perfect Accuracy, Precision, Recall, and F1score values, which are better than 99.92%, but its MCC values are greater than 99.77%. Three imbalanced data handling approaches can enhance the model performance of models trained by three algorithms. Moreover, the experiment of reducing the number of features based on mutual information calculation revealed slightly decreasing performance for 50 data features with Accuracy values greater than 99.86%. For 25 data features, which is the smallest size, the random forest supervised model yielded 99.15% Accuracy. Both sampling strategies and feature selection help to improve the supervised model for accurately predicting credit risk, which may be beneficial in the lending business.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Enhancing Supervised Model Performance in Credit Risk Classification Using Sampling Strategies and Feature Ranking

Wattanakitrungroj,

Wijitkajee,

Jaiyen

et al. 2024

BDCC

View full text Add to dashboard Cite

show abstract

“…For performing data categorization with good accuracy, K-nearest neighbor (K-NN), decision tree, support vector machine, and logistic regression models are taken into account to measure their performance. A loan default dataset was used in [25], which is taken from the lending club. To address the dataset's class imbalance issue, the ADASYN (Adaptive Synthetic Sampling Approach) method was used in increasing the prediction accuracy.…”

Section: A Ml-based Loan Predictionmentioning

confidence: 99%

Eligible Personal Loan Applicant Selection using Federated Machine Learning Algorithm

Anannya,

Khatun,

Hosen

et al. 2023

IJACSA

View full text Add to dashboard Cite

Loan sanctioning develops a paramount financial dependency amongst banks and customers. Banks assess bundles of documents from individuals or business entities seeking loans depending on different loan types since only reliable candidates are chosen for the loan. This reliability materializes after assessing the previous transaction history, financial stability, and other diverse kinds of criteria to justify the reliance of the bank on an applicant. To reduce the workload of this laborious assessment, in this research, a machine learning (ML) based web application has been initiated to predict eligible candidates considering multiple criteria that banks generally use in their calculation, in short which can be briefed as loan eligibility prediction. Data from prior customers, who are authorized for loans based on a set of criteria, are used in this research. As ML techniques, Random Forest, K-Nearest Neighbour, Adaboost, Extreme Gradient Boost Classifier, and Artificial Neural Network algorithms are utilized for training and testing the dataset. A federated learning approach is employed to ensure the privacy of loan applicants. Performance analysis reveals that Random Forest classifier has provided the best output with an accuracy of 91%. Based on the mentioned prediction, the web application can decide whether the customers' requested loan should be accepted or rejected. The application was developed using NodeJs, ReactJS, Rest API, HTML, and CSS. Furthermore, parameter tuning can improve the performance of the web application in the future along with a usable user interface ensuring global accessibility for various types of users.

show abstract

“…These advanced methodologies not only offer higher accuracy but also provide nuanced insights into the complex dynamics of loan default risks. Moreover, the CART decision-tree model [3], integration of LSTM neural networks [4] and a blend of algorithms that combine logistic regression, Random Forest, and CatBoost [5] has emerged, leveraging the strengths of novel models.…”

Section: Introductionmentioning

confidence: 99%

Machine Learning-Based Loan Default Prediction in Peer-to-Peer Lending

Yang

2024

HSET

View full text Add to dashboard Cite

The peer-to-peer (P2P) lending market has recently undergone significant growth, transforming traditional lending practices. However, this evolution brings with it unique challenges, particularly in managing credit risk and ensuring the reliability of loan approvals. Accurate prediction of loan defaults remains a pivotal aspect of risk management in this sector. This study introduces a comprehensive approach to improve bad loan prediction in peer-to-peer (P2P) lending, sourcing Lending Club data. In the face of challenges posed by imbalanced datasets and risk management in the loan industry, our methodology significantly enhances prediction accuracy, particularly in identifying bad loans. The study implements a comprehensive process that includes data cleaning, feature engineering, feature selection, balancing the dataset and machine learning models, achieving a noteworthy accuracy rate over 92% and recall rate above 87%. This research advances academic understanding of loan prediction and generates real-world impact. The accurate models identifying explanatory provide a valuable framework for improving decision-making and strategic planning in P2P lending platforms.

show abstract

Prediction of loan default based on multi-model fusion

Cited by 10 publications

References 14 publications

Enhancing Supervised Model Performance in Credit Risk Classification Using Sampling Strategies and Feature Ranking

Enhancing Supervised Model Performance in Credit Risk Classification Using Sampling Strategies and Feature Ranking

Eligible Personal Loan Applicant Selection using Federated Machine Learning Algorithm

Machine Learning-Based Loan Default Prediction in Peer-to-Peer Lending

Contact Info

Product

Resources

About