Academic Journals Database
Disseminating quality controlled scientific knowledge

Bankruptcy Prediction using SVM and Hybrid SVM Survey

ADD TO MY LIST
 
Author(s): Jayanthi. J | Suresh Joseph. K | Vaishnavi. J

Journal: International Journal of Computer Applications
ISSN 0975-8887

Volume: 34;
Issue: 7;
Start page: 39;
Date: 2011;
Original page

Keywords: Bankruptcy prediction | SVM

ABSTRACT
Bankruptcy prediction has been a topic of active research for business and corporate organizations since past decades. It is an effective tool to help financial institutions and relevant people to make the right decision in investments, especially in the current competitive environment. The tool provides auditors and managers a chance to identify the problems early. Thus, relevant people have an opportunity to intervene early into affairs problems to reduce the expected cost of bankruptcy failure. The primary objective is to reduce well in advance as much as possible to the loss of relevant beneficial, such as investors, managers and employees. This paper provides a survey of bankruptcy predictions using a novel machine learning approach: support vector machine 'SVM' and hybrid SVM. The underlying motivation for using SVM and hybrid SVM are the ability of this methodology to determine accurate forecast bankruptcy data set when the underlying system processes are typically nonlinear, nonstationary and not defined a - priori. SVM and hybrid SVM have also been proven to outperform other nonlinear techniques including neuralnetwork based nonlinear prediction techniques.
Save time & money - Smart Internet Solutions     

Tango Jona
Tangokurs Rapperswil-Jona