A new deep learning ensemble credit risk evaluation model with an improved synthetic minority oversampling technique

Shen, Feng; Zhao, Xingchao; Kou, Gang; Alsaadi, Fawaz E.

doi:10.1016/j.asoc.2020.106852

Cited by 123 publications

(57 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Plawiak et al (2020) proposed a Deep Genetic Hierarchical Network of Learners (DGHNL) credit scoring model integrating SVM, KNN, probabilistic neural network, and fuzzy system, and the validity of the method was proved by German credit data set in UCI database [ 30 ]. To deal with the imbalance of credit data, Shen et al (2021) developed a new deep learning ensemble credit risk assessment model that combined the LSTM algorithm and the AdaBoost algorithm, and compared the performance of the proposed model and other widely used credit scoring models on two imbalanced credit data sets [ 31 ].…”

Section: Literature Reviewmentioning

confidence: 99%

“…Although the existing studies have begun to focus on combining multiple feature selection methods to improve the performance of the classifier, there is a lack of interpretable optimal feature set determination method analysis. Besides, existing studies have found that in the field of default discrimination, deep learning can reveal the complex relationship between credit data variables, making its performance better than traditional statistical methods and machine learning methods [ 31 ]. However, existing research on deep learning models still has shortcomings such as many hyperparameters, requiring a large amount of training data, and determination the structure of the neural network before training.…”

Section: Literature Reviewmentioning

confidence: 99%

See 1 more Smart Citation

A Two-Stage Hybrid Default Discriminant Model Based on Deep Forest

Liu

et al. 2021

Entropy

View full text Add to dashboard Cite

Background: the credit scoring model is an effective tool for banks and other financial institutions to distinguish potential default borrowers. The credit scoring model represented by machine learning methods such as deep learning performs well in terms of the accuracy of default discrimination, but the model itself also has many shortcomings such as many hyperparameters and large dependence on big data. There is still a lot of room to improve its interpretability and robustness. Methods: the deep forest or multi-Grained Cascade Forest (gcForest) is a decision tree depth model based on the random forest algorithm. Using multidimensional scanning and cascading processing, gcForest can effectively identify and process high-dimensional feature information. At the same time, gcForest has fewer hyperparameters and has strong robustness. So, this paper constructs a two-stage hybrid default discrimination model based on multiple feature selection methods and gcForest algorithm, and at the same time, it optimizes the parameters for the lowest type II error as the first principle, and the highest AUC and accuracy as the second and third principles. GcForest can not only reflect the advantages of traditional statistical models in terms of interpretability and robustness but also take into account the advantages of deep learning models in terms of accuracy. Results: the validity of the hybrid default discrimination model is verified by three real open credit data sets of Australian, Japanese, and German in the UCI database. Conclusion: the performance of the gcForest is better than the current popular single classifiers such as ANN, and the common ensemble classifiers such as LightGBM, and CNNs in type II error, AUC, and accuracy. Besides, in comparison with other similar research results, the robustness and effectiveness of this model are further verified.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Section: Literature Reviewmentioning

confidence: 99%

A Two-Stage Hybrid Default Discriminant Model Based on Deep Forest

Liu

et al. 2021

Entropy

View full text Add to dashboard Cite

show abstract

“…e system is mainly used to assess the aesthetic quality of professional color photographs. Shen et al [23] refer to [24] for predicting the aesthetic description of book covers: "good," "bad," or "ugly." However, they did not propose a method to improve the correct rate of classification prediction.…”

Section: Related Workmentioning

confidence: 99%

Research on Modern Book Packaging Design under Aesthetic Evaluation Based on Deep Learning Model

2021

Discrete Dynamics in Nature and Society

View full text Add to dashboard Cite

The development of information technology has led to the rapid development of modern book packaging design. Book packaging design is different from painting. It is a kind of design process that combines artistry and practicality and has double characteristics. With the continuous progress of society, people’s requirements for book packaging design have become higher and higher, and modern book packaging design has become an important topic in the field of art design. To this end, this paper introduces the machine learning algorithms used in this paper, including the AdaBoost algorithm and the SVR algorithm. Specifically, it includes the principles and specific implementation steps for AdaBoost classification algorithm and SVR regression algorithm, as well as evaluation indexes of AdaBoost classification and SVR regression analysis. Realization of physical books reflects artistry, creativity, professionalism, popularity, and vitality of books in the packaging and design of books in the electronic information era. The stimulation effect of this paper starts from packaging design, inspection mechanism, and brand psychology to get the superior design in modern book packaging design.

show abstract

“…The authors focused on parametrizing GAMs for the task at hand and then applied GAMs to the dataset while finding that the methodology is more predictively accurate on out‐of‐sample sets. Shen et al 12 studied a credit evaluation model based on an ensemble algorithm. The authors developed an enhanced SMOTE variant, following which the long‐ short‐term memory (LSTM) network and adaptive boosting (AdaBoost) were implemented.…”

Section: Background and Previous Workmentioning

confidence: 99%

A Gaussian process‐based approach toward credit risk modeling using stationary activations

Mahajan

Nayyar

Raina

et al. 2021

Concurrency and Computation

View full text Add to dashboard Cite

The task of predicting the risk of defaulting of a lender using tools in the domain of AI is an emerging one and in growing demand, given the revolutionary potential of AI. Various attributes like income, properties acquired, educational status, and many other socioeconomic factors can be used to train a model to predict the possibilities of nonrepayment of a loan or its chances. Most of the techniques and algorithms used in this regard previously do not submit any attention to the uncertainty in predictions for out of distribution (OOD) in a dataset, which contributes to overfitting, leading to relatively lower accuracy for predicting these data points. Specifically, for credit risk classification, this is a serious concern, given the structure of the available datasets and the trend they follow. With a focus on this issue, we propose a robust and better methodology that uses a recent and efficient family of nonlinear neural network activation functions, which mimics the properties induced by the widely‐used Matérn family of kernels in Gaussian process (GP) models. We tested the classification performance metrics on three openly available datasets after prior preprocessing. We achieved a high mean classification accuracy of 87.4% and a lower mean negative log predictive density loss of 0.405.

show abstract

A new deep learning ensemble credit risk evaluation model with an improved synthetic minority oversampling technique

Cited by 123 publications

References 34 publications

A Two-Stage Hybrid Default Discriminant Model Based on Deep Forest

A Two-Stage Hybrid Default Discriminant Model Based on Deep Forest

Research on Modern Book Packaging Design under Aesthetic Evaluation Based on Deep Learning Model

A Gaussian process‐based approach toward credit risk modeling using stationary activations

Contact Info

Product

Resources

About