A hybrid predictive framework for evaluating P2P credit risks

Abstract

PurposeDespite better accessibility and flexibility, peer-to-peer (P2P) lending has suffered from excessive credit risks, which may cause significant losses to the lenders and even lead to the collapse of P2P platforms. The purpose of this research is to construct a hybrid predictive framework that integrates classification, feature selection, and data balance algorithms to cope with the high-dimensional and imbalanced nature of P2P credit data.Design/methodology/approachAn improved synthetic minority over-sampling technique (IMSMOTE) is developed to incorporate the randomness and probability into the traditional synthetic minority over-sampling technique (SMOTE) to enhance the quality of synthetic samples and the controllability of synthetic processes. IMSMOTE is then implemented along with the grey relational clustering (GRC) and the support vector machine (SVM) to facilitate a comprehensive assessment of the P2P credit risks. To enhance the associativity and functionality of the algorithm, a dynamic selection approach is integrated with GRC and then fed in the SVM's process of parameter adaptive adjustment to select the optimal critical value. A quantitative model is constructed to recognize key criteria via multidimensional representativeness.FindingsA series of experiments based on real-world P2P data from Prosper Funding LLC demonstrates that our proposed model outperforms other existing approaches. It is also confirmed that the grey-based GRC approach with dynamic selection succeeds in reducing data dimensions, selecting a critical value, identifying key criteria, and IMSMOTE can efficiently handle the imbalanced data.Originality/valueThe grey-based machine-learning framework proposed in this work can be practically implemented by P2P platforms in predicting the borrowers' credit risks. The dynamic selection approach makes the first attempt in the literature to select a critical value and indicate key criteria in a dynamic, visual and quantitative manner.

A novel quality risk evaluation framework for complex equipment development integrating PHFS-QFD and grey clustering

“…The grey clustering model is one of the evaluation methods and has a wide range of applications (Delcea et al. , 2022; He et al. , 2022; Karimi and Yahyazade, 2022; Li et al.…”

“…It stands out in handling situations with inadequate information and dealing with uncertain problems Hu and Liu, 2022;Liu et al, 2022a, b;Tao et al, 2022). The grey clustering model is one of the evaluation methods and has a wide range of applications (Delcea et al, 2022;He et al, 2022;Karimi and Yahyazade, 2022;Li et al, 2022;Luo et al, 2022;Liu et al, 2022a, b) In this stage, a grey clustering model based on mixed end-point possibility function is applied to evaluate the quality risk of systems. This model is suitable for various types of problems with clear boundaries, but points that are most likely to belong to each gray class are not clear (Liu et al, 2015).…”

Machine Learning and Artificial Intelligence Method for FinTech Credit Scoring and Risk Management

“…They have addressed these technologies' impact, adoption, and potential to reshape the financial sector, emphasizing their relevance to sustainability, customer experience, and responsible AI adoption. Several studies are valuable for policymakers, businesses, and researchers navigating the complex terrain of FinTech, technology, and sustainability (e.g., Ahmed et al, 2022;Goodell et al, 2021;He et al, 2022;Kumar & Kaur 2023;Mahmud et al, 2023;Martinelli et al, 2020;Najem et al, 2022;Noreen et al, 2023;Shaik, 2023;Weber et al, 2023).…”

Enhancing Default Prediction in P2P Lending using Random Forest and Grey Wolf Optimization-based Feature Selection

“…The Max-Relevance and Min-Redundancy (MRMR) technique is employed for feature selection, and k-means clustering aids in discarding irrelevant attributes [1]. Various models, including LightGBM (Light Gradient Boosting Machine) [21], Random Forest [22][23], Logistic Regression [24], Random Forest and XGBoost [24], Binary Particle Swarm Optimization (PSO) with Support Vector Machines (SVM) [25], Adaptive Feature Selection based on Most Informative Graph and Most Relative Graph [26], and Grey Relational Clustering, have demonstrated the superior accuracy of results achieved through feature selection compared to unselected features [27].…”