Improving Credit Risk Prediction in Online Peer-to-Peer (P2P) Lending Using Imbalanced Learning Techniques

Ferreira, Luís; Barddal, Jean Paul; Gomes, Heitor Murilo; Enembreck, Fabrício

doi:10.1109/ictai.2017.00037

Cited by 18 publications

(15 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…3, May 2021 Extreme Learning Machine (ELM), incremental extreme learning machine (I-ELM), and support vector machine(SVM), by using the dataset of CRE and then discussed its advantages and disadvantages. [16] compared three single classifiers with the model they proposed on the Lending Club dataset to analyze the influence of supervised classification models and unbalanced data processing technique to credit prediction rates. [17] proposed Mahalanobis distance induced kernels in support vector machines with application in CRE and compared with traditional SVM kernels.…”

Section: A Dbm-drbm Hybrid Modelmentioning

confidence: 99%

A Credit Risk Predicting Hybrid Model Based on Deep Learning Technology

Wu¹

2021

IJMLC

View full text Add to dashboard Cite

Credit risk evaluation (CRE) is a very challenging and important management science problem in the domain of financial analysis. Many popular methods have been applied to tackle this problem in recent years. However, feature extraction and imbalanced data problem have not been paid enough attention in the current research, which play significant function in field of CRE. In this paper, we employed a deep learning approach to extract effective features and under-sampling technique to balance dataset. Our model combine under-sampling technique, Deep Boltzmann Machine (DBM) and Discriminative Restricted Boltzmann Machine (DRBM) method. To examine the performance, real world credit data of Lending Club is applied in the proposed model. The stable and better performance results show that the Hybrid classifier we propose is more effective and powerful.

show abstract

Section: A Dbm-drbm Hybrid Modelmentioning

confidence: 99%

A Credit Risk Predicting Hybrid Model Based on Deep Learning Technology

Wu¹

2021

IJMLC

View full text Add to dashboard Cite

show abstract

“…Machine learning has been also used to address important aspects at a predictive level such as the problem of class imbalance [33], [34] in the optimization of hyperparameters, for example, using genetic algorithms [17] and in feature selection [35] or the improvement of bias reflected in the prediction [36], [37].…”

Section: Machine Learning and Credit Risk Modeling In P2p Lendingmentioning

confidence: 99%

“…The sampling method used in the second case is a hybrid between undersampling and oversampling [67], also used in [33], [34], which consists of equating the minority class proportion, default, with the nondefault randomly selected cases from the majority class and generating random records from the minority class.…”

Section: B Model Adjustment 1) Logistic Regressionmentioning

confidence: 99%

See 1 more Smart Citation

Explainability of a Machine Learning Granting Scoring Model in Peer-to-Peer Lending

Garzón

Arroyo

Caparrini³

et al. 2020

IEEE Access

View full text Add to dashboard Cite

Peer-to-peer (P2P) lending demands effective and explainable credit risk models. Typical machine learning algorithms offer high prediction performance, but most of them lack explanatory power. However, this deficiency can be solved with the help of the explainability tools proposed in the last few years, such as the SHAP values. In this work, we assess the well-known logistic regression model and several machine learning algorithms for granting scoring in P2P lending. The comparison reveals that the machine learning alternative is superior in terms of not only classification performance but also explainability. More precisely, the SHAP values reveal that machine learning algorithms can reflect dispersion, nonlinearity and structural breaks in the relationships between each feature and the target variable. Our results demonstrate that is possible to have machine learning credit scoring models be both accurate and transparent. Such models provide the trust that the industry, regulators and end-users demand in P2P lending and may lead to a wider adoption of machine learning in this and other risk assessment applications where explainability is required.

show abstract

“…With the wide application of machine learning, online learning with concept drift and class imbalance has received increased research attention. Practical applications in software engineering, risk management, traffic flows, sensor networks and social media mining face challenges of both concept drift and class imbalance [1], [2].…”

Section: Introductionmentioning

confidence: 99%

Resample-Based Ensemble Framework for Drifting Imbalanced Data Streams

et al. 2019

View full text Add to dashboard Cite

Machine learning in real-world scenarios is often challenged by concept drift and class imbalance. This paper proposes a Resample-based Ensemble Framework for Drifting Imbalanced Stream (RE-DI). The ensemble framework consists of a long-term static classifier to handle gradual and multiple dynamic classifiers to handle sudden concept drift. The weights of the ensemble classifier are adjusted from two aspects. First, a time-decayed strategy decreases the weights of the dynamic classifiers to make the ensemble classifier focus more on the new concept of the data stream. Second, a novel reinforcement mechanism is proposed to increase the weights of the base classifiers that perform better on the minority class and decrease the weights of the classifiers that perform worse. A resampling buffer is used for storing the instances of the minority class to balance the imbalanced distribution over time. In our experiment, we compare the proposed method with other state-of-the-art algorithms on both real-world and synthetic data streams. The results show that the proposed method achieves the best performance in terms of both Prequential AUC and accuracy.INDEX TERMS Online ensemble learning, resample learning, reinforcement, concept drift, class imbalance.

show abstract

Improving Credit Risk Prediction in Online Peer-to-Peer (P2P) Lending Using Imbalanced Learning Techniques

Cited by 18 publications

References 35 publications

A Credit Risk Predicting Hybrid Model Based on Deep Learning Technology

A Credit Risk Predicting Hybrid Model Based on Deep Learning Technology

Explainability of a Machine Learning Granting Scoring Model in Peer-to-Peer Lending

Resample-Based Ensemble Framework for Drifting Imbalanced Data Streams

Contact Info

Product

Resources

About