A novel heterogeneous ensemble credit scoring model based on bstacking approach

Xia, Yufei; Liu, Chuanzhe; Da, Bowen; Xie, Fangming

doi:10.1016/j.eswa.2017.10.022

Cited by 178 publications

(98 citation statements)

References 75 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The two main types of ensemble models are homogeneous (combining same classifiers) and heterogeneous (combining different classifiers). In credit scoring domain, [56][57][58] worked on homogeneous ensembles while Xia et al [59] worked on heterogeneous ensembles. Improve Predictive Ability.…”

Section: Simultaneous Hyperparameters Tuning and Features Selectionmentioning

confidence: 99%

Credit Scoring: A Review on Support Vector Machines and Metaheuristic Approaches

Goh

Lee

2019

Advances in Operations Research

View full text Add to dashboard Cite

Development of credit scoring models is important for financial institutions to identify defaulters and nondefaulters when making credit granting decisions. In recent years, artificial intelligence (AI) techniques have shown successful performance in credit scoring. Support Vector Machines and metaheuristic approaches have constantly received attention from researchers in establishing new credit models. In this paper, two AI techniques are reviewed with detailed discussions on credit scoring models built from both methods since 1997 to 2018. The main discussions are based on two main aspects which are model type with issues addressed and assessment procedures. Then, together with the compilation of past experiments results on common datasets, hybrid modelling is the state-of-the-art approach for both methods. Some possible research gaps for future research are identified.

show abstract

Section: Simultaneous Hyperparameters Tuning and Features Selectionmentioning

confidence: 99%

Credit Scoring: A Review on Support Vector Machines and Metaheuristic Approaches

Goh

Lee

2019

Advances in Operations Research

View full text Add to dashboard Cite

show abstract

“…The model was validated using five real world credit scoring datasets to demonstrate its ability to improve classification performance against all base classifiers. Xia, Liu, Da, and Xie () propose a novel heterogeneous ensemble credit model that integrates the bagging algorithm with the stacking method. The proposed model differs from the extant ensemble credit models in three aspects, namely, pool generation, selection of base learners, and trainable fuser.…”

Section: Literature Reviewmentioning

confidence: 99%

“…Several comparative experiments between some state‐of‐the‐art ensemble credit scoring methods will also conducted to further clarify the efficiency of the proposed model. They are Consensus approach proposed in Ala'Raj and Abbod (), Heterogeneous ensemble credit model proposed in Xia et al (), EBCA‐RF&XGB‐PSO proposed in He et al (), and dynamic ensemble classification method based on soft probability proposed in Feng et al (). Table displays the final results of these comparison models in different datasets.…”

Section: Experimental Analysismentioning

confidence: 99%

Generalized fuzzy soft sets theory‐based novel hybrid ensemble credit scoring model

Zhang

Feng

2018

Int. J Fin Econ

View full text Add to dashboard Cite

In banking and peer‐to‐peer loan applicant firms, customer credit scores have numerous applications in risk control and precision marketing. Numerous credit scoring techniques act as classification methods. In this paper, the main issue is simultaneous and hybrid utilization of the feature selection (FS) algorithm and ensemble learning classification algorithms with respect to their parameter settings to achieve higher performance in the proposed credit scoring model. As a result, this paper reports a hybrid data mining model of generalized fuzzy soft sets (GFSS) theory‐based ensemble learning classification algorithms based on three stages. The first stage addresses data gathering and preprocessing. The second stage uses the adaptive elastic net‐based FS algorithm to eliminate irrelevant or weakly correlated variables. After the appropriate variables are chosen, they are applied to the proposed ensemble model. In the third stage, GFSS theory is exploited to construct a novel weight assignment mechanism for each individual credit scoring model in the ensemble model according to performance. Several comparisons are conducted to validate the proposed model using real world credit datasets. The experimental results, analysis, and statistical tests prove the ability of the proposed approach to improve classification performance against all of the base classifier, hybrid, and combination methods in terms of average accuracy, area under the curve, H‐measure, and Brier score.

show abstract

“…Most of the current credit scoring methods are based on statistical approaches which consider multi-dimension attribute information about the applicant of the loan. For instance, the logistic regression model, the ordered probit model, artificial neural network (ANN) algorithms, support vector machines (SVM) are widely used for predicting the probability of default for clients (Byanjankar et al 2015, Xia et al 2018, Ignatius et al 2018, Wang et al 2019b). While it is not always the "best" model in credit scoring problem (West 2000).…”

Section: Introductionmentioning

confidence: 99%

Credit scoring by incorporating dynamic networked information

Wang

Djehiche

et al. 2020

European Journal of Operational Research

View full text Add to dashboard Cite

In this paper, the credit scoring problem is studied by incorporating network information, where the advantages of such incorporation are investigated in two scenarios. Firstly, a Bayesian optimal filter is proposed to provide a prediction for lenders assuming that published credit scores are estimated merely from structured individual data. Such prediction is used as a monitoring indicator for the risk warning in lenders' future financial decisions. Secondly, we further propose a recursive Bayes estimator to improve the accuracy of credit scoring estimation by incorporating the dynamic interaction topology of clients as well. It is shown that under the proposed evolution framework, the designed estimator has a higher precision than any efficient estimator, and the mean square errors are strictly smaller than the Cramér-Rao lower bound for clients within a certain range of scores. Finally, simulation results for a specific case illustrate the effectiveness and feasibility of the proposed methods.

show abstract

A novel heterogeneous ensemble credit scoring model based on bstacking approach

Cited by 178 publications

References 75 publications

Credit Scoring: A Review on Support Vector Machines and Metaheuristic Approaches

Credit Scoring: A Review on Support Vector Machines and Metaheuristic Approaches

Generalized fuzzy soft sets theory‐based novel hybrid ensemble credit scoring model

Credit scoring by incorporating dynamic networked information

Contact Info

Product

Resources

About