Hybrid CRO Based FLANN for Financial Credit Risk Forecasting
Modern financial market has become capable enough to provide its services to a large number of customers simultaneously. On the other hand, the exponential hike in financial crises per year has uplifted the demand for precise and potential classifier models. In this work a hybrid model of clustering and neural network based classifier has been proposed, i.e. FCM-FLANN-CRO. Three financial credit risk data sets were applied to the processing and the model is evaluated using the performance metrics such as RMSE and Accuracy. The experimental result shows the proposed model outperforms its MLP counterpart and other two non-hybrid models. The proposed model provides its best result with 97.05 % of classification accuracy.
KeywordsClassification Functional link artificial neural network (FLANN) Multilayered perceptron (MLP) Credit risk forecasting Chemical reaction optimization (CRO) Clustering Fuzzy C-means (FCM)
- 2.Public Company Bankruptcy Filing Report, New Generation Research, Inc. Available from http://www.bankruptcydata.com/default.asp
- 6.Altman, E.I., Avery, R.B., Eisenbeis, R.A., Sinkey Jr, J.F.: Application of classification techniques in business: banking and finance. JAI Press Inc., CT (1981)Google Scholar
- 10.Kennedy, J., Eberhart, R.: Particle swarm optimization. In: Proceedings of 1995 IEEE International Conference on Neural Networks, vol. 4, pp. 1942–1948 (1995)Google Scholar