This research is aimed at establishing the diagnosis models for business crises through integrating a real-valued genetic algorithm to determine the optimum parameters and SVM to perform learning and classification on data. After finishing the training processes, the proposed GA-SVM can reach a prediction accuracy of up to 95.56% for all the tested business data. Particularly, only six influential features are included in the proposed model with intellectual capital and financial features after the 2-phase selecting process; the six features are ordinary and widely available from public business reports. The proposed GA-SVM is available for business managers to conduct self-diagnosis in order to realize whether business units are really facing a crisis. © 2007 Elsevier Ltd. All rights reserved.
Chen, L. H., & Hsiao, H. D. (2008). Feature selection to diagnose a business crisis by using a real GA-based support vector machine: An empirical study. Expert Systems with Applications, 35(3), 1145–1155. https://doi.org/10.1016/j.eswa.2007.08.010