• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 1999.03a
• /
• pp.415-421
• /
• 1999

Altman의 연구(1965, 1977)나 Beaver의 연구(1986)와 같은 전통적 예측모형은 분석자의 판단에 따른 예측도가 높은 재무비율을 선정하여 다변량판별분석(MDA:multiple discriminant analysis), 로지스틱회귀분석 등과 같은 통계기법을 주로 이용해 왔으나 1980년 후반부터 인공지능 기법인 귀납적 학습방법, 인공신경망모형, 유전모형 등이 부실기업예측에 응용되기 시작했다. 최근 연구에서는 인공신경망을 활용한 변수 및 모형개발에 관한 보고가 있다. 그러나 지금까지의 연구가 주로 기업의 재무적 비율지표를 고려한 모형에 치중되었으며 정성적 자료인 비재무지표에 대한 검증과 선정이 자의적으로 이루어져온 경향이었다. 또한 너무 많은 입력변수를 사용할 경우 다중공선성 문제를 유발시킬 위험을 내포하고 있다. 본 연구에서는 부실기업예측모형을 수립하기 위하여 정량적 요인인 재무적 지표변수와 정성적 요인인 비재무적 지표변수를 모두 고려하였다. 재무적 지표변수는 상관분석 및 요인분석들을 통하여 유의한 변수들을 도출하였으며 비재무적 지표변수는 조직생태학내에서의 조직군내 조직사멸과 관련된 생태적 과정에 대한 요인들 중 조직군 내적요인으로 조직의 연령, 조직의 규모, 조직의 산업밀도를 도출하여 4개의 실험집단으로 분류하여 비재무적 지표변수를 보완하였다. 인공신경망은 다층퍼셉트론(multi-layer perceptrons)과 역방향 학습(back-propagation)알고리듬으로 입력변수와 출력변수, 그리고 하나의 은닉층을 가지는 3층 퍼셉트론(three layer perceptron)을 사용하였으며 은닉층의 노드(node)수는 3개를 사용하였다. 입력변수로 안정성, 활동성, 수익성, 성장성을 나타내는 재무적 지표변수와 조직규모, 조직연령, 그 조직이 속한 산업의 밀도를 비재무적 지표변수로 산정하여 로지스틱회귀 분석과 인공신경망 기법으로 검증하였다. 로지스틱회귀분석 결과에서는 재무적 지표변수 모형의 전체적 예측적중률이 87.50%인 반면에 재무/비재무적 지표모형은 90.18%로서 비재무적 지표변수 사용에 대한 개선의 효과가 나타났다. 표본기업들을 훈련과 시험용으로 구분하여 분석한 결과는 전체적으로 재무/비재무적 지표를 고려한 인공신경망기법의 예측적중률이 높은 것으로 나타났다. 즉, 로지스틱회귀 분석의 재무적 지표모형은 훈련, 시험용이 84.45%, 85.10%인 반면, 재무/비재무적 지표모형은 84.45%, 85.08%로서 거의 동일한 예측적중률을 가졌으나 인공신경망기법 분석에서는 재무적 지표모형이 92.23%, 85.10%인 반면, 재무/비재무적 지표모형에서는 91.12%, 88.06%로서 향상된 예측적중률을 나타내었다.

• Journal of the Korean Geotechnical Society
• /
• v.24 no.1
• /
• pp.119-130
• /
• 2008

Uncertainties in physical and engineering parameters for the design of shallow foundations arise from various aspects such as inherent variability and measurement error. This paper aims at investigating and reducing uncertainty from deterministic method by using the reliability-based design of shallow foundations accounting for the variation of various design parameters. A probability distribution type and statistics of random variables such as unit weight, cohesion, infernal friction angle and Young's modulus in geotechnical engineering are suggested to calculate the ultimate bearing capacities and immediate settlements of foundations. Reliability index and probability of failure are estimated based on the distribution types of random variables. Widths of foundation are calculated at target reliability index and probability of failure. It is found that application and analysis of the best-fit distribution type for each random variables are more effective than adoption of the normal distribution type in optimizing the reliability-based design of shallow foundations.

• 한국노년학
• /
• v.32 no.3
• /
• pp.777-800
• /
• 2012

This paper aims to estimate the effects of environmental variables on volunteering among older persons and decide relationships between independent and dependent variables. The thesis conceptually points out that the integrated theory of resources too much emphasizes the important roles of human, social and cultural capital, but overlooks the influences of ecological environments in explaining volunteering among the older persons. And the thesis tries to apply the general ecological theory of aging to explaining volunteering of older people together with resource frameworks, and to estimate the effects of ecological environment variables on volunteerism for senior citizens. Using a micro data of 2009 National Social Survey by Statistics Korea, the paper screens out 10,268 subjects who are believed to socially retire and be above 55 years older. The multiple OLS regression and binomial logistic regression techniques are used to estimate the effects of ecological environments and resources on volunteering. The analysis results show that all of environmental and resource variables are related to volunteering at the level of p<.000. This means that environmental variables have independent effects on the volunteerism, controlling for resource variables. This results suggest that both theories have empirical evidences in explaining volunteerism in Korea. Also, at the end of paper, theoretical and policy implications for practices and future studies are discussed.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.15 no.1
• /
• pp.1-15
• /
• 2016

This study aims to develop a model by which city bus drivers who are likely to cause an accident can be figured out based on the information about their actual driving records. For this purpose, from the information about the actual driving records of the drivers who have caused an accident and those who have not caused any, significance variables related to traffic accidents are drawn, and the accuracy between models is compared for the classification models developed, applying a discriminant analysis and logistic regression analysis. In addition, the developed models are applied to the data on other drivers' driving records to verify the accuracy of the models. As a result of developing a model for the classification of drivers who are likely to cause an accident, when deceleration ($X_{deceleration}$) and acceleration to the right ($Y_{right}$) are simultaneously in action, this variable was drawn as the optimal factor variable of the classification of drivers who had caused an accident, and the prediction model by discriminant analysis classified drivers who had caused an accident at a rate up to 62.8%, and the prediction model by logistic regression analysis could classify those who had caused an accident at a rate up to 76.7%. In addition, as a result of the verification of model predictive power of the models showed an accuracy rate of 84.1%.

• Information Systems Review
• /
• v.17 no.1
• /
• pp.49-64
• /
• 2015

Customer reviews which include subjective opinions for the product or service in online store have been generated rapidly and their influence on customers has become immense due to the widespread usage of SNS. In addition, a number of studies have focused on opinion mining to analyze the positive and negative opinions and get a better solution for customer support and sales. It is very important to select the key terms which reflected the customers' sentiment on the reviews for opinion mining. We proposed a document-level terms-based sentiment classification model by select in the optimal terms with part of speech tag. SVMs (Support vector machines) are utilized to build a predictor for opinion mining and we used the combination of POS tag and four terms extraction methods for the feature selection of SVM. To validate the proposed opinion mining model, we applied it to the customer reviews on Amazon. We eliminated the unmeaning terms known as the stopwords and extracted the useful terms by using part of speech tagging approach after crawling 80,000 reviews. The extracted terms gained from document frequency, TF-IDF, information gain, chi-squared statistic were ranked and 20 ranked terms were used to the feature of SVM model. Our experimental results show that the performance of SVM model with four POS tags is superior to the benchmarked model, which are built by extracting only adjective terms. In addition, the SVM model based on Chi-squared statistic for opinion mining shows the most superior performance among SVM models with 4 different kinds of terms extraction method. Our proposed opinion mining model is expected to improve customer service and gain competitive advantage in online store.

• Journal of the Korea Convergence Society
• /
• v.6 no.5
• /
• pp.1-8
• /
• 2015

The data of fair value hierarchy is expected to contain different degree of measurement error, information asymmetry, and information risk by the level of hierarchy. Thus, this study examines how hierarchy of fair value discriminately influences on companies' cost of capital. Through regression analysis of corporations listed from 2011 to 2014, it turns out that the regression coefficient of level 1 and 2 of fair value variable vary their rank by cost of capital types, while level 3 contains the highest regression coefficient for every cost of capital variable. In addition, further study of how the relevance between cost of capital and the fair value hierarchy gets affected by market risk management level and audit quality finds no consistent results. However, by analyzing the effect of coincident interaction through the convergence of market risk management and audit quality, when audit quality and market risk management level are high, the effect of relieving cost of capital of Level 3 gets the highest. In conclusion, fair value hierarchy data seems to affect discriminately on cost of capital by involved information risk, and the information risk could decrease by the level of market risk management and audit quality.

• Geotechnical Engineering
• /
• v.9 no.3
• /
• pp.77-90
• /
• 1993

In the settlement analysis of shallow foundation soil properties, loads and soil strata involve many uncertainties so it is necessary to do analysis of settlement that considers the probabilistic properties of each variable. This study is performed to probabilistic analysis for settlement of shallow foundation consisted of individual footings by using Monte Carlo Method. To consider the uncertainty of variables, both the soil properties and loads are assumed to be normal distribution random variables and get settlement mean and coefficient of variation of individual footing. And the settlement of each individual footing is also assumed to be normal distribution. Settlement of each individual footing which considers the probability of soft soil pockets in soil strata follows Markov process. Then it is performed to do sensitivity analysis which is involved to excess probability of allowable criteria of maxi mum settlement and differential settlement according to varity of each variable. It is thought to be proper that the settlement analysis of shallow foundation should be analyzed considering uncertainty of variables and soil stratum conditions.

• Journal of the Society of Naval Architects of Korea
• /
• v.28 no.2
• /
• pp.241-250
• /
• 1991

The reliability analysis for web frame of tanker is carried out by the probabilistic finite element method combined with the classical reliability method such as MVFOSM and AFOSM which can be used for calculating the probability of failure for the complicated structures in which the limit state equation is implicitly expressed. As random variables external load, elastic modulus, sectional moment of inertia and field stress are chosen and Parkinson's iteration algorithm in AFOSM is used for reliability analysis. By adding only the covariance data of the random variables to the input data set required for conventional finite element method, the present method can easily calculate the probability of failure at every element end as well as the covariances of structural reponses such as displacements at every element end and member forces at every element, even for the complicated ship structure.

• Computational Structural Engineering
• /
• v.4 no.2
• /
• pp.111-117
• /
• 1991

The analysis method calculating the mean and standard deviation for the eigenvalue of complicated structures in which the limit state equation is implicitly expressed is formulated and applied to the buckling analysis by combining probabilistic finite element method with direct differential method which is a kind of sensitivity analysis technique. Also, the probability of buckling failure is calculated by combining classical reliability techniques such a MVFOSM and AFOSM. As random variables external load, elastic modulus, sectional moment of inertia and member length are chosen and Parkinson's iteration algorithm in AFOSM is used. The accuracy of the results by this study is verified by comparing the results with the crude Monte Carlo simulation and Importance Sampling Method. Through the case study of some structures the important aspects of buckling reliability analysis are discussed.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.11 no.4
• /
• pp.91-102
• /
• 1991

Traditionally, a dynamic programming(DP) technique has been used to the multireservoir control system. The algorithm has inherent problem to increase computational requirements exponentially due to discretization of variables and expanding the dimension of the system. To solve this problem, this paper describes transforming the optimal control system into a quadratic programming(QP), using Laguerre polynomials(LP) and its properties. The objective function of the proposed QP is independent of time variable. The solution of the QP is obtained by nonlinear programming(NLP) using augmented Lagrangian multiplier method. The numerical experiment shows that the water level of reservoirs is higher than Lee's and the evaluated benefit value is about the same as other researcher's.