• Journal of the Korean Operations Research and Management Science Society
• /
• v.21 no.1
• /
• pp.81-100
• /
• 1996

본 연구에서는 보다 효과적인 기업신용평가를 위하여, 통계적 방법과 인공지능 방법을 결합한 결합모형을 제시햐고자 한다. 이를 위하여 본 연ㄴ구에서는 통계적인 모형중에서 가장 널리 활용되고 있는 MDA (Multivariate Discriminant Analysis) 와 인공지능적인 방법으로서 최근에 널리 사용되고 있는 인공싱경망( neural network)모형을 휴리스틱한 방법으로 결합한다. 이러한 결합모형의 성과를 증명하기 위하여 우리나라의 대표적인 3대 기업신용평가 기관에서 수집한 1043개의 기업신용평가자료를 기초로 실혐을 하고, 그 결과를 기존의 MDA 및 인공신경망 방법에 의한 결과와 비교하였다. 실험결과, 통계적으로도 유의하고, 실무적인 관점에서도 의미가 있는 기업신용펑가 결과를 유도할 수 있었다.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1996.10a
• /
• pp.214-217
• /
• 1996

Early decision support systems (DSS) were the "passive" decision support systems in the sense that the systems only able to do what the users explicitly direct them to do. But some researchers such as Raghav Rao et al. [51 showed architectures to suggest general idea of the innovative DSS systems which offer active form of decision support, say, "active Intelligent Decision Support Systems(active IDSS)". The system can perform not only what the users want to do but some voluntary (or involuntary) intelligent works. This paper presents the issues in the design of the active IDSS in the domain of investment evaluation, a domain area where few researchers have suggested frameworks or architectures to discriminate good investment from bad one. We propose a new paradigm, by utilizing historical investment results using neural network and Multivariate Discriminant Analysis(MDA), to identify goodness of investment. A new active IDSS architecture which consists of neural network, expert system and three components of the traditional passive DSS is suggested with some scenario based results.nario based results.

• Proceedings of the KOSOMBE Conference
• /
• v.1992 no.05
• /
• pp.155-160
• /
• 1992

Liver function test is one of the most common tests for diagnosis and follow-up of patients and for heal th screening. Automatic interpretation and suggestions on the diagnostic possibilities contribute to shorten the interpretation time of the test results and help to provide qualified health care. Fuzzy logic has been recently introduced and being spread for these purposes. The present study aims at model Ins the foray rule-based laboratory diagnosis system. The fuzzy rule-based laboratory diagnosis system was applied to the diagnosis regarding liver function test. The system was evaluated by comparing with the stepwise multivariate discriminant function analysis, which showed similar results, and the overall accuracy of the fuzzy diagnosis system was about 80%.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.18 no.2
• /
• pp.57-81
• /
• 1993

This paper is concerned with analyzing the bankruptcy prediction power of three methods : Multivariate Discriminant Analysis (MDA), Inductive Learning, Neural Network, MDA has been famous for its effectiveness for predicting bankrupcy in accounting fields. However, it requires rigorous statistical assumptions, so that violating one of the assumptions may result in biased outputs. In this respect, we alternatively propose the use of two AI models for bankrupcy prediction-inductive learning and neural network. To compare the performance of those two AI models with that of MDA, we have performed massive experiments with a number of Korean bankrupt-cases. Experimental results show that AI models proposed in this study can yield more robust and generalizing bankrupcy prediction than the conventional MDA can do.

• Asia pacific journal of information systems
• /
• v.5 no.1
• /
• pp.36-51
• /
• 1995

본 논문에서는 비구조적인 의사결정문제를 효과적으로 해결하기 위하여 감독학습 인공신경망 모형과 비감독학습 인공신경망 모형을 결합한 하이브리드 인공신경망 모형인 HYNEN(HYbrid NEural Network) 모형을 제안한다. HYNEN모형은 주어진 자료를 클러스터화 하는 CNN(Clustering Neural Network)과 최종적인 출력을 제공하는 ONN(Output Neural Network)의 2단계로 구성되어 있다. 먼저 CNN에서는 주어진 자료로부터 적정한 퍼지규칙을 찾기 위하여 클러스터를 구성한다. 그리고 이러한 클러스터를 지식베이스로하여 ONN에서 최종적인 의사결정을 한다. CNN에서는 SOFM(Self Organizing Feature Map)과 LVQ(Learning Vector Quantization)를 클러스터를 만든 후 역전파학습 인공신경망 모형으로 이를 학습한다. ONN에서는 역전파학습 인공신경망 모형을 이용하여 각 클러스터의 내용을 학습한다. 제안된 HYNEN 모형을 우리나라 기업의 도산자료에 적용하여 그 결과를 다변량 판별분석법(MDA:Multivariate Discriminant Analysis)과 ACLS(Analog Concept Learning System) 퍼지 ARTMAP 그리고 기존의 역전파학습 인공신경망에 의한 실험결과와 비교하였다.

• Food Science of Animal Resources
• /
• v.32 no.5
• /
• pp.669-677
• /
• 2012

The objective of this study was to analyze the flavor pattern of different varieties of cheeses. Four of the each following cheese varieties such as shred type pizza cheese, Cheddar cheese, Mozzarella block cheese, and white mold-ripened cheeses, stored at $4^{\circ}C$ during 2 wks were examined before and after cooking at $70^{\circ}C$ and $160^{\circ}C$. Flavor patterns of these cheeses were analyzed using an electronic nose system based on mass spectrometer. All data were treated by multivariate data processing based on discriminant function analysis (DFA). The results showed the discriminant model by DFA method. Data revealed that flavor patterns of pizza cheeses were well separated as storage prolonged and obviously discriminated as the higher the cooking temperature. The result of pattern recognition analysis based on discriminant function analysis showed that new brand of pizza cheese produced by Imsil Cheese Cooperative was located at middle between the flavors of the imported brands of pizza cheese and those of domestic brand of pizza cheeses. Imsil cheese has a unique flavor pattern among other variety of cheeses. Application of pattern recognition analysis by electronic nose might be useful and advanced technology for characterizing in flavor pattern of cheese products from different origins and different categories of cheeses.

• The Research Journal of the Costume Culture
• /
• v.8 no.1
• /
• pp.30-39
• /
• 2000

The purpose of this study were to 1) identify types and levels of production environments, 2) classify apparel manufacturers based on production environments and 3) investigate relationship between characteristics of apparel manufacturers and production environment. Apparel manufacturer's characteristics included product line and the number of employees. For this study, the questionnaires were administered to 215 apparel manufacturers in seoul and Kyung-gi region from Feb. to Mar. 1998. Employing a sample of 201, data were analyzed by factor analysis, descriptive statistics, cluster analysis, cluster analysis, discriminant Analysis, and multivariate analysis of variance. The following are the results of this study : 1. The production environment was identified as three types such as complexity of product environment, uncertainty of demand/supply environment and uncertainty of worker environment. 2. Based on three types of the production environment, apparel manufacturers were classified into stable group, uncertain group and complicated group. 3. With respect to product line, men's wear manufacturers were lied the most high complexity of product environment, casual wear and knit wear were lied the most frequently uncertainty of worker environment. With respect to the number employees, apparel manufacturers comprising 50∼99 employees were lied the most high complexity of product environment, while those comprising 100∼299 employees the most high demand/supply environment.

• The Korean Journal of Applied Statistics
• /
• v.11 no.2
• /
• pp.435-449
• /
• 1998

In this article we have introduced a Bayes criterion for the variable selection in multiple discriminant analysis (MDA). The criterion is a default Bayes factor for the comparision of homo/heteroscadasticity of the multivariate normal means. The default Bayes factor is obtained from a development of the imaginary training sample method introduced by Spiegelhalter and Smith (1982). Based an the criterion, we also provided a test for additional discrimination in MDA. The advantage of the criterion is that it is not only applicable for the optimal subset selection method but for the stepwise method. More over, the criterion can be reduced to that for two-group discriminant analysis. Thus the criterion can be regarded as an unified alternative to variable selection criteria suggested by various sampling theory approaches. To illustrate the performance of the criterion, a numerical study has bean done via Monte Carlo experiment.

• The International Journal of Costume Culture
• /
• v.3 no.3
• /
• pp.235-245
• /
• 2000

The purposes of this study were 1) to investigate usage level of production strategies based on group of production environment, 2) to investigate usage level of production systems based on group of production strategy, and 3) to analyze each of company performance based on group of production strategy and system. For this study, the questionnaires were administered to 215 apparel manufactures in metropolitan area from Feb. to Mar. 1998. Employing a sample of 201, data were analyzed by factor analysis, descriptive statistics, cluster analysis, discriminant analysis, and multivariate analysis of variance. The following are the results of this study. 1. Concerning production strategy due to group of production environment, the stable group and the complicated group prefer to rice/quality centered strategy but the level of usage for strategies is so pretty that it is not significant to carry out them. 2. Concerning production system due to group of production strategy, the workers centered group is occupied high in the price/quality centered group & the complex group. And also the product centered system is occupied high in the flexibility centered group. 3. Concerning company performance due to group of production strategy and system, the price/quality centered group holds low position of performance comparing to another groups. And the performance of the managers centered group is higher than that of the workers.

• Korean journal of food and cookery science
• /
• v.20 no.4
• /
• pp.371-379
• /
• 2004

The purpose of this study was to identify factors associated with high satisfaction with foodservices in family restaurant. Accordingly, this study surveyed questionnaire concerning 20 measures of foodservice as well as major subject descriptors. The result of this study were as follows. KMO and Bartlett's test statistics showed that the data fitted factor analysis well. Results of factor analysis, average variance extracted estimates and shared variance showed that the convergent and discriminant validitys of 3 factors are supported, and cronbach's alpha showed that the internal consistency of 3 factors is supported. It was found that expensive groups, except gender groups and frequency of purchase groups, were differentially associated with high levels of overall satisfaction with foodservices. Multivariate analyses showed that satisfaction with service factor was the best predictor of overall satisfaction, followed by facilities factor. Three factors emerged from the logistic regression analysis as predictors of level of overall satisfaction. Overall, approximately 77% of university students could be correctly classified as being satisfied or unsatisfied. Finally, the results of the study provide some insights into the market-oriented types of foodservice marketing strategies or tactics to enable family restaurant to effectively manage and more competitive.