Please use this identifier to cite or link to this item:
|Title:||A systematic credit scoring model based on heterogeneous classifier ensembles|
|Keywords:||Science & Technology;Technology;Computer Science, Information Systems;Computer Science, Interdisciplinary Applications;Computer Science;Credit scoring;LR;ANN;SVM;Homogenous ensembles;Heterogeneous ensembles;Bagging;Majority voting;Support Vector Machines;Neural-Networks;Bankruptcy Prediction;Risk-assessment;Trees|
|Citation:||International Symposium on Innovations in Intelligent Systems and Applications (INISTA) Proceedings, pp. 119 - 125, (2015)|
|Abstract:||Lending loans to borrowers is considered one of the main profit sources for banks and financial institutions. Thus, careful assessment and evaluation should be taken when deciding to grant credit to potential borrowers. With the rapid growth of credit industry and the massive volume of financial data, developing effective credit scoring models is very crucial. The literature in this area is very dense with models that aim to get the best predictive performance. Recent studies stressed on using ensemble models or multiple classifiers over single ones to solve credit scoring problems. Therefore, this study propose to develop and introduce a systematic credit scoring model based on homogenous and heterogeneous classifier ensembles based on three state-of-the art classifiers: logistic regression (LR), artificial neural network (ANN) and support vector machines (SVM). Results revealed that heterogeneous classifier ensembles gives better predictive performance than homogenous and single classifiers in terms of average accuracy.|
|Appears in Collections:||Dept of Electronic and Computer Engineering Research Papers|
Items in BURA are protected by copyright, with all rights reserved, unless otherwise indicated.