Corporate Credit Rating based on Bankruptcy Probability Using AdaBoost Algorithm-based Support Vector Machine

Taeksoo Shin(신택수),Taeho Hong(홍태호) 한국지능정보시스템학회 2011 지능정보연구 Vol.17 No.3

Recently, support vector machines (SVMs) are being recognized as competitive tools as compared with other data mining techniques for solving pattern recognition or classification decision problems. Furthermore, many researches, in particular, have proved them more powerful than traditional artificial neural networks (ANNs) (Amendolia et al., 2003; Huang et al., 2004, Huang et al., 2005; Tay and Cao, 2001; Min and Lee, 2005; Shin et al., 2005; Kim, 2003).The classification decision, such as a binary or multi-class decision problem, used by any classifier, i.e. data mining techniques is so cost-sensitive particularly in financial classification problems such as the credit ratings that if the credit ratings are misclassified, a terrible economic loss for investors or financial decision makers may happen. Therefore, it is necessary to convert the outputs of the classifier into wellcalibrated posterior probabilities-based multiclass credit ratings according to the bankruptcy probabilities. However, SVMs basically do not provide such probabilities. So it required to use any method to create the probabilities (Platt, 1999; Drish, 2001). This paper applied AdaBoost algorithm-based support vector machines (SVMs) into a bankruptcy prediction as a binary classification problem for the IT companies in Korea and then performed the multi-class credit ratings of the companies by making a normal distribution shape of posterior bankruptcy probabilities from the loss functions extracted from the SVMs. Our proposed approach also showed that their methods can minimize the misclassification problems by adjusting the credit grade interval ranges on condition that each credit grade for credit loan borrowers has its own credit risk, i.e. bankruptcy probability.

몽골 시장에서 한류에 대한 태도가 한국의 이미지와 제품신뢰, 구전 및 제품 선호도에 미치는 영향

체렌돌람(Tserendulam),신택수(Taeksoo Shin) 대한경영학회 2014 대한경영학회 학술발표대회 발표논문집 Vol.2014 No.2

An Integrated Model of SNA and AI Methods for Predicting Serendipitous Items

Tserendulam(체렌돌람),Taeksoo Shin(신택수) 한국지능정보시스템학회 2018 한국지능정보시스템학회 학술대회논문집 Vol.2018 No.11

원주시 건강도시 웹 데이터베이스 구축

남은우(Eun Woo Nam),신택수(Taeksoo Shin),송예리아(Yea-Li-A Song),박재성(Jae Sung Park),박기수(Ki Soo Park),송태민(Tae Min Song),김민경(Min Kyung Kim) 한국보건교육건강증진학회 2007 보건교육건강증진학회지 Vol.24 No.3

SVM과 meta-learning algorithm을 이용한 고지혈증 유병 예측모형 개발과 활용

이슬기(Seulki Lee),신택수(Taeksoo Shin) 한국지능정보시스템학회 2018 지능정보연구 Vol.24 No.2

This study aims to develop a classification model for predicting the occurrence of hyperlipidemia, one of the chronic diseases. Prior studies applying data mining techniques for predicting disease can be classified into a model design study for predicting cardiovascular disease and a study comparing disease prediction research results. In the case of foreign literatures, studies predicting cardiovascular disease were predominant in predicting disease using data mining techniques. Although domestic studies were not much different from those of foreign countries, studies focusing on hypertension and diabetes were mainly conducted. Since hypertension and diabetes as well as chronic diseases, hyperlipidemia, are also of high importance, this study selected hyperlipidemia as the disease to be analyzed. We also developed a model for predicting hyperlipidemia using SVM and meta learning algorithms, which are already known to have excellent predictive power. In order to achieve the purpose of this study, we used data set from Korea Health Panel 2012. The Korean Health Panel produces basic data on the level of health expenditure, health level and health behavior, and has conducted an annual survey since 2008. In this study, 1,088 patients with hyperlipidemia were randomly selected from the hospitalized, outpatient, emergency, and chronic disease data of the Korean Health Panel in 2012, and 1,088 nonpatients were also randomly extracted. A total of 2,176 people were selected for the study. Three methods were used to select input variables for predicting hyperlipidemia. First, stepwise method was performed using logistic regression. Among the 17 variables, the categorical variables(except for length of smoking) are expressed as dummy variables, which are assumed to be separate variables on the basis of the reference group, and these variables were analyzed. Six variables (age, BMI, education level, marital status, smoking status, gender) excluding income level and smoking period were selected based on significance level 0.1. Second, C4.5 as a decision tree algorithm is used. The significant input variables were age, smoking status, and education level. Finally, C4.5 as a decision tree algorithm is used. In SVM, the input variables selected by genetic algorithms consisted of 6 variables such as age, marital status, education level, economic activity, smoking period, and physical activity status, and the input variables selected by genetic algorithms in artificial neural network consist of 3 variables such as age, marital status, and education level. Based on the selected parameters, we compared SVM, meta learning algorithm and other prediction models for hyperlipidemia patients, and compared the classification performances using TP rate and precision. The main results of the analysis are as follows. First, the accuracy of the SVM was 88.4% and the accuracy of the artificial neural network was 86.7%. Second, the accuracy of classification models using the selected input variables through stepwise method was slightly higher than that of classification models using the whole variables. Third, the precision of artificial neural network was higher than that of SVM when only three variables as input variables were selected by decision trees. As a result of classification models based on the input variables selected through the genetic algorithm, classification accuracy of SVM was 88.5% and that of artificial neural network was 87.9%. Finally, this study indicated that stacking as the meta learning algorithm proposed in this study, has the best performance when it uses the predicted outputs of SVM and MLP as input variables of SVM, which is a meta classifier. The purpose of this study was to predict hyperlipidemia, one of the representative chronic diseases. To do this, we used SVM and meta-learning algorithms, which is known to have high accuracy. As a result, the accuracy of classification of hyperlipidemia in the stac

전통시장 상인회의 조직특성이 사회적 자본과 상인회 조직성과에 미치는 영향

김민숙 ( Min Sook Kim ),신택수 ( Taeksoo Shin ) 한국지식경영학회 2016 지식경영연구 Vol.17 No.4

Korean traditional markets have been struggling of late as big-sized superstores and SSM(Super Supermarkets) are thriving in the market. They have therefore upgraded their facilities and undertaken management modernization actively to overcome the threat to traditional markets and ensure their competitiveness; however, the effect does not appear to be verifiable. The purpose of this study is to analyze the impact of the organizational characteristics of the traditional market merchant association on social capital and organizational performance. In other words, this paper investigates a merchant association`s organizational characteristics in terms of the modernization of business activities of the traditional markets and the influence on their social capital and organizational performance. This study analyzes the traditional market by evaluating the impact of these factors. This study consists of four hypotheses: The first hypothesis relates to the causal relationship between the characteristics of a merchant association and social capital. The second and third hypotheses, respectively, relate to the causal relationships between the social capital of a merchant association and the merchant`s satisfaction and that between the social capital of a merchant association and organizational commitment. The last hypothesis relates to the relationship between the organizational commitment of a merchant association and the merchant`s satisfaction. This study conducts a reliability and validity analysis of the above factors and analyzes the causal relationships between them by using the PLS(Partial Least Squares) path model as one of the structural equation models. The results of the empirical analysis are summarized as follows: First, the organizational characteristics of the traditional market merchant association have a significant influence on social capital. However, only two sub-hypotheses are not significant; these insignificant hypotheses relate to the relationship between a merchant`s entrepreneurship and structural capital and that between a merchant`s entrepreneurship and cognitive capital. Second, the social capital of a merchant association influences organizational commitment significantly. Third, the relationship between the social capital of a merchant association and the merchant`s satisfaction is mostly significant. However, one of the sub-hypotheses, that is, the relationship between relational capital and a merchant`s satisfaction is not exceptionally significant. Lastly, the organizational commitment of a merchant association affect the merchant`s satisfaction significantly. Through our extensive study, this paper found that a merchant association`s organizational characteristics of the traditional market significantly affect social capital, organizational commitment, and satisfaction through the mediation of social capital. Therefore, in order to activate the key traditional market, an understanding of organizational characteristics and social capital is primarily required. Systematic management and investment pertaining to these two factors will be the first consideration for revitalizing traditional markets.

소각재와 폐콘크리트를 이용한 재생골재의 고형화 특성

연익준(Yeon Ikjun),주소영(Ju Soyoung),이상우(Lee Sangwoo),신택수(Shin Taeksoo),김광렬(Kim Kwangyul) 한국지반환경공학회 2008 한국지반환경공학회논문집 Vol.9 No.5

본 연구에서는 각종 건축물의 해체 공사에서 발생하는 다량의 폐콘크리트를 이용한 재생골재의 활용방안으로 유해폐기물고형화 방법에 있어 요구되는 재생골재의 공학적 특성을 파악하고 그 적합성을 평가하기 위하여 실험하였다. 시멘트 모르타르에 재생골재를 기존 모래와 혼합하여 재생골재 5-15%일때 28일 양생 모르타르의 강도는 1급 콘크리트 벽돌 C종의 강도기준인 163kgf/㎠을 모두 상회하였다. 안정성 평가를 위한 용출실험 결과 Cu, Cd, Pb, Cr, As의 경우 법적기준치 보다 낮았으나, Hg의 경우 기준치보다 용출농도가 높게 나타내었는데 이는 Hg이 단순히 물리적으로 고정되기 때문인 것으로 판단된다. 시멘트의 모르타르의 결정구조를 파악하기 위한 XRD 분석결과 Ca(OH)₂, ettringite, CSH 피크가 나타났으며, 양생시간이 길어짐에 따라 나타나는 비교적 안정된 수화물인 CSH의 피크가 높게 나타났다 위의 실험 결과 재생골재 치환량 5-10%일 때 혼화재인 비산회, 하수슬러지 소각재를 첨가한 경우 유해폐기물고형화 재료로써 활용가능성이 충분하나 압축강도를 좀 더 증진시킬 수 있는 다른 혼화재 및 첨가제 등에 관한 연구가 진행된다면 효과적인 재활용 방안이 될 것으로 판단된다. In this study, It was carried out to evaluate the feasibility of recycled crushed concrete as aggregate used cement mortar replace sand and to investigate engineering properties of recycled aggregate for hazardous waste solidification. The compressive strength of cement mortar replaced 5-15% (wt.) recycled aggregate was over 163 kgf/㎠ which is the standard of first grade concrete block class C. And cement mortar was examined to evaluate the stability by leaching test. Cu, Cd, Pb, Cr, and As as the heavy metals were proved very stable but mercury (Hg) was leached high concentration because it was simply tied to the cement surface. We investigated the crystal structures of cement mortar and they had shown the peaks of Ca(OH)₂, ettringite, and CSH (calcium silicate hydrate). As the result, the longer curing time, the higher CSH peak that means to increase compressive strength and the cement mortar was more stable. Therefore it was shown that it may be possible to apply hazardous waste solidification using recycled aggregate, fly ash and sewage sludge ash.

Utilization of Forecasting Accounting Earnings Using Artificial Neural Networks and Case-based Reasoning : Case Study on Manufacturing and Banking Industry 제조업과 은행업을 중심으로

최용석,한인구,신택수 한국경영과학회 2003 한국경영과학회지 Vol.28 No.3

The financial statements purpose to provide useful information to decision-making process of business managers. The value-relevant information. however, embedded in the financial statement has been often overlooked in Korea. In fact the financial statements in Korea are been utilized for nothing but account reports to Security Supervision Boards (SSB). The objective of this study is to develop earnings forecasting models through financial statement analysis using artificial intelligence (AI). AI methods are employed in forecasting earnings : artificial neural networks (ANN) for manufacturing industry and case-based reasoning (CBR) for banking industry. The experimental results using such AI methods are as follows. Using ANN for manufacturing industry records 63.2% of hit ratio for out-of-sample, which outperforms the logistic regression by around 4%. The experiment through CBR for banking industry shows 65.0% of hit ratio that beats the statistical method by 13.2% in holdout sample. Finally, the prediction results for manufacturing industry are validated through monitoring g the shift in cumulative returns of portfolios based on the earning prediction. The portfolio wit the firms whose earnings are predicted to increase is designated as best portfolio and the portfolio wit the earnings-decreasing firms as worst portfolio. The difference between tow portfolios is about 3% of cumulative abnormal return on average. Consequently, this result showed that the financial statements In Korea contain the value-relevant information that is not reflected in stock prices.