http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
판별분석, 규칙기반예측, 신경회로망을 이용한 도산예측의 비교 분석
조홍규,한인구,이훈영 한국전문가시스템학회 1994 학술대회 Vol.2 No.1
Bankruptcy prediction is a major area of business classification problem. This paper compared the prediction accuracy by three methods: multivariate discriminant analysis, case-based forecasting, and neural network. This is the first study to compare the performance of analogical reasoning and neural network. The data used in the experiment are selected from Korean bankrupt and non-bankrupt firms for the recent three years. The three models are used in this paper. Multivariate discriminant analysis (MDA) is a traditional statistical model. The case based forecasting system (CBFS) is composed of the three sub-processes: determining key attributes in identifying similar cases to predict the target variables, accessing similarity and retrieving analogous cases, and generating a forecast through combining similar cases selected. The neural network (NN) model adopted here is back propagation. The hit ratio, the rate of correct prediction is not significantly different between the MDA and CBFS, while NN is superior to the other two methods. The average hit ratios of three methods are ranged from 81.5% to 83.8%, but the best hit ratios of MDA, CBFS, and neural network are 88.8%, 87.3%, and 90.9%.
A Review of Artificial Intelligence Models in Business Classification
한인구,권영식,조홍규 한국지능정보시스템학회 1995 지능정보연구 Vol.1 No.1
Business researchers have traditionally used statistical techniques for classification. In late 1980's, inductive learning started to be used for business classification. Recently, neural network began to be applied for business classification. This study reviews the business classification studies, identifies a neural network approach as the most powerful classification tool, and discusses the problems and issues in neural network applications.