[PDF] Credit Risk Evaluation in Peer-to-peer Lending With Linguistic Data Transformation and Supervised Learning | Semantic Scholar (2024)

A system was built using the CatBoost method to detect credit risk for minimizing P2P lending credit risk and the best result regard to scenario 2 with a data splitting of 90:10 was caused by the result of AUC value’s 0.80329.

This article compared two learning algorithms – Logistic regression and Artificial Neural Network to classify borrowers based on loan repayment schedule and revealed that both approaches were robust in classifying late borrowers with logistic regression being 0.02% more robust than Neural Network.

The paper presents evaluation scales of imprecise phrases commonly used during the process of credit risk assessment based on experts’ preferences and uses a scoring function determined with the use of an adapted Simple Additive Weighting (SAW) method.

The Default Borrowers Detection Model (DefBDet) is a novel model, it can predict a loan status based on a multi-classification bases rather than a binary class bases, so that special conditions are assigned before loan being approved.

The findings show that social media and e-WOM strategies play a critical role and have a positive significant impact on a CF campaign.

The proposed P2P Lending for Tourism platform in the Indonesian context consists of business process and application, which are adopted from the p2p lending and combine with e-commerce.

...

...

This study uses data from the Lending Club to explore the characteristics of loan and its applicant and uses random forest to do the feature selection in the modeling phase, and shows that the term of loan, annual income, the amount of loans, debt-to-income ratio, credit grade and revolving line utilization play an important role in loan defaults.

A credit scoring model using artificial neural networks in classifying peer-to-peer loan applications into default and non-default groups is proposed and results indicate that the neural network-basedcredit scoring model performs effectively in screening default applications.

A multi relational Bayesian classification method is used to predict the default probabilities of borrowers, where the input attributes consists of both core credit and social network information.

Experimental studies using real world data sets have demonstrated that the classification and regression trees and neural networks outperform the traditional credit scoring models in terms of predictive accuracy and type II errors.

This paper presents the current achievements and limitations associated with the development of bankruptcy-prediction and credit-scoring models employing machine learning, and provides suggestions for future research.

Based on BP neural network, P2P network platform credit risk evaluation model were built to train and simulate and the results show that the final data generated by these models are of practical value.

...

...