Developing a business failure prediction model via RST, GRA and CBR

and CBR Rong-Ho Lin , Chun-Lin g Chuang, Kuan-Wei Huang National Taipei University of Technology, Institute of Commerce Automation & Management No. 1, Section 3, Chung-Hsiao East Road, Taipei, 106, Taiwan, ROC Kainan University, Department of Information Management & Graduate No. 1, Kainan Road, Luzhu, Taoyuan,33857 Taiwan, ROC * [email protected] Abstract The prediction of business failure is an important and challenging issue that has served as the impetus for many academic studies over the past three decades. The widely applied methods to predict the risk of business failure were the classic statistical methods, data mining and machine learning techniques. Case Based-Reasoning (CBR) is an inductive machine learning method that can apply to diagnosis domain, classification, and enhanced some of the deficiencies in statistical models. Concerning attributes extraction and weighting approach could enable CBR to retrieve the most similar case correctly and effectively. This paper proposes a hybrid prediction failure (HFP) model by applying Rough Set Theory (RST) and Grey Relational Analysis (GRA) as data preprocessors to strength the effectiveness of CBR predicting capability. After exploring TEJ database and comparing with three models, the results show that our model is the most accurate and effective in predicting business failure.