Title: Support vector machine and wavelet neural network hybrid: application to bankruptcy prediction in banks
Authors: Devulapalli Karthik Chandra, Vadlamani Ravi, Pediredla Ravisankar
Addresses: Institute for Development and Research in Banking Technology, Castle Hills Road #1, Masab Tank, Hyderabad 500057 (AP), India. ' Institute for Development and Research in Banking Technology, Castle Hills Road #1, Masab Tank, Hyderabad 500057 (AP), India. ' Institute for Development and Research in Banking Technology, Castle Hills Road #1, Masab Tank, Hyderabad 500057 (AP), India
Abstract: This paper presents a novel soft computing system, SVWNN, to predict failure of banks. First, support vectors that are critical in classification are extracted from support vector machine (SVM). Then, these support vectors along with their corresponding actual output labels are used to train the wavelet neural network (WNN). Further, Garson|s algorithm for feature selection is adapted using WNN. Thus, the new hybrid, WNN-SVWNN, accomplishes horizontal and vertical reduction in the dataset as support vectors reduce the pattern space dimension and the WNN-based feature selection reduces the feature space dimension. The effectiveness of these hybrids is demonstrated on the datasets of US, Turkish, UK and Spanish banks. SVWNN outperformed SVM and WNN on all datasets except Spanish banks. However, when feature selection is considered, WNN-SVM outperformed WNN-WNN and WNN-SVWNN on Spanish and Turkish banks, while WNN-SVWNN outscored others on UK banks. Ten-fold cross-validation was performed throughout the study.
Keywords: bankruptcy prediction; support vector machines; SVM; wavelet neural networks; WNNs; feature selection; support vectors; data mining; soft computing; bank failure; banking industry; US banks; USA; United States; Turkey; Turkish banks; UK banks; United Kingdom; Spain; Spanish banks.
International Journal of Data Mining, Modelling and Management, 2010 Vol.2 No.1, pp.1 - 21
Published online: 18 Jan 2010 *Full-text access for editors Access for subscribers Purchase this article Comment on this article