• Journal of Intelligence and Information Systems
• /
• v.18 no.2
• /
• pp.29-45
• /
• 2012

Bond rating is regarded as an important event for measuring financial risk of companies and for determining the investment returns of investors. As a result, it has been a popular research topic for researchers to predict companies' credit ratings by applying statistical and machine learning techniques. The statistical techniques, including multiple regression, multiple discriminant analysis (MDA), logistic models (LOGIT), and probit analysis, have been traditionally used in bond rating. However, one major drawback is that it should be based on strict assumptions. Such strict assumptions include linearity, normality, independence among predictor variables and pre-existing functional forms relating the criterion variablesand the predictor variables. Those strict assumptions of traditional statistics have limited their application to the real world. Machine learning techniques also used in bond rating prediction models include decision trees (DT), neural networks (NN), and Support Vector Machine (SVM). Especially, SVM is recognized as a new and promising classification and regression analysis method. SVM learns a separating hyperplane that can maximize the margin between two categories. SVM is simple enough to be analyzed mathematical, and leads to high performance in practical applications. SVM implements the structuralrisk minimization principle and searches to minimize an upper bound of the generalization error. In addition, the solution of SVM may be a global optimum and thus, overfitting is unlikely to occur with SVM. In addition, SVM does not require too many data sample for training since it builds prediction models by only using some representative sample near the boundaries called support vectors. A number of experimental researches have indicated that SVM has been successfully applied in a variety of pattern recognition fields. However, there are three major drawbacks that can be potential causes for degrading SVM's performance. First, SVM is originally proposed for solving binary-class classification problems. Methods for combining SVMs for multi-class classification such as One-Against-One, One-Against-All have been proposed, but they do not improve the performance in multi-class classification problem as much as SVM for binary-class classification. Second, approximation algorithms (e.g. decomposition methods, sequential minimal optimization algorithm) could be used for effective multi-class computation to reduce computation time, but it could deteriorate classification performance. Third, the difficulty in multi-class prediction problems is in data imbalance problem that can occur when the number of instances in one class greatly outnumbers the number of instances in the other class. Such data sets often cause a default classifier to be built due to skewed boundary and thus the reduction in the classification accuracy of such a classifier. SVM ensemble learning is one of machine learning methods to cope with the above drawbacks. Ensemble learning is a method for improving the performance of classification and prediction algorithms. AdaBoost is one of the widely used ensemble learning techniques. It constructs a composite classifier by sequentially training classifiers while increasing weight on the misclassified observations through iterations. The observations that are incorrectly predicted by previous classifiers are chosen more often than examples that are correctly predicted. Thus Boosting attempts to produce new classifiers that are better able to predict examples for which the current ensemble's performance is poor. In this way, it can reinforce the training of the misclassified observations of the minority class. This paper proposes a multiclass Geometric Mean-based Boosting (MGM-Boost) to resolve multiclass prediction problem. Since MGM-Boost introduces the notion of geometric mean into AdaBoost, it can perform learning process considering the geometric mean-based accuracy and errors of multiclass. This study applies MGM-Boost to the real-world bond rating case for Korean companies to examine the feasibility of MGM-Boost. 10-fold cross validations for threetimes with different random seeds are performed in order to ensure that the comparison among three different classifiers does not happen by chance. For each of 10-fold cross validation, the entire data set is first partitioned into tenequal-sized sets, and then each set is in turn used as the test set while the classifier trains on the other nine sets. That is, cross-validated folds have been tested independently of each algorithm. Through these steps, we have obtained the results for classifiers on each of the 30 experiments. In the comparison of arithmetic mean-based prediction accuracy between individual classifiers, MGM-Boost (52.95%) shows higher prediction accuracy than both AdaBoost (51.69%) and SVM (49.47%). MGM-Boost (28.12%) also shows the higher prediction accuracy than AdaBoost (24.65%) and SVM (15.42%)in terms of geometric mean-based prediction accuracy. T-test is used to examine whether the performance of each classifiers for 30 folds is significantly different. The results indicate that performance of MGM-Boost is significantly different from AdaBoost and SVM classifiers at 1% level. These results mean that MGM-Boost can provide robust and stable solutions to multi-classproblems such as bond rating.

• Clinical and Experimental Pediatrics
• /
• v.51 no.1
• /
• pp.1-25
• /
• 2008

Purpose : Since 1967, The Korean Pediatric Society and Korean Government have developed Korean Growth Standards every 10 years. Last version was published in 1998. During past 40 years, Korean Growth Standards were mainly descriptive charts without any systematic nor statistical standardization. With the global epidemic of obesity, many authorities such as World Health Organization (WHO) and United States' Centers for Disease Control (CDC) have been changed their principles of growth charts to cope with the situations like ours. This article summarizes and reviews the whole developmental process of new 2007 Korean Growth Charts with discussion. Methods : With the initiative of Division of Chronic Disease Surveillance in Korea Centers for Disease Control and Prevention, we have performed new national survey for the development of new Standards in 2005 and identified marked increase of childhood obesity and plateau of secular increment of final height in late adolescents. We have developed new Growth Standards via adapting several innovative methods, including standardization of all available raw data, which were acquired in 1997 and 2005 national survey and full application of LMS method. Results : We could get new standardized charts; weight-for-age, length/height-for-age, weight-for-height, head circumference-for-age and BMI-for-age. Other non-standardized charts based on 2005 survey data were also published; waist circumference-for-age, mid-arm circumference-for-age, chest circumference-for-age and skinfold-for-age. Clinical guideline was also developed. Conclusion : Developmental process and results of new Korean Growth Charts are comparable with other internationally well-known Growth Standards, WHO 2006 Growth Standards and CDC Growth Charts. 2007 Korean Growth Charts are relevant especially in Korea and Korean ethnic groups.

• Journal of the Economic Geographical Society of Korea
• /
• v.14 no.4
• /
• pp.467-485
• /
• 2011

After Korean War, Korea focused on catching up with the world economy by concentrating on some target industries around the Capital Region and southern coastal cities. Thus, the regional disparity between Capital Region and non-Capital Regions increased drastically. At last, when Korea acquired full-fledged autonomy in 1994 in the Civilian government (1993-1998) and experienced the Asian financial crisis in 1997-1998, local governments were awakened to the notion of region-oriented development, especially for regional industrial development. The purposes of this paper are to introduce regional industrial development policies since 1998 and to suggest some recommendations in terms of how to adjust regional development for industrial policies in the future. In the introducing phase (Kim administration, 1998-2003), four provincial governments requested national funding to raise regional industries that are of strategic importance. At the same time, the central government recognized the need to nurture regional industries to overcome structural weaknesses. As a result, the Roh administration (2003-2008) gave a birth to a systematizing phase. As the ultimate regional policy objective, the balanced national development has been set and the Special Acts, Special Accounts, Committee, and National Plan have been established. Regional Industrial Promotion Project has been carried out very actively during this period. It had a good start albeit idealistic to a certain extent. Therefore, the current government has changed policy paradigm from balanced growth to regional competitiveness along with global paradigm shifts. In order to enhance regional competitiveness, regional development policies have been pursued in more efficient way. Leading Industry Nurturing Projects (LINPs) on Economic Region level, existed Regional Industrial Promotion Projects (RIPPs) on Province level, and Region Specific Industry Projects (RSIPs) on Local Area level have been implemented. Now, it is appropriate to review regional development policies including industrial policies since 1998 and to adjust them for the future sustainable regional development. Because LINPs and RIPPs will be terminated in next two years, the 2nd stage projects are on planning to reduce the redundancies in two projects. In addition, business support program would be reformed from subsiding technology development to building ecological business system. Finally some policy implications are provided in this paper, which is useful to establish the new regional industrial policies for both central and local government.

• Korean Journal of Plant Resources
• /
• v.30 no.2
• /
• pp.213-218
• /
• 2017

Rice is the staple food of at least half of the world's population. Due to global warming, the weather is difficult to forecast nowadays. Therefore, it is necessary to breed various breeding to respond to such changes in the environment. This study was conducted to analyze the QTL about plant form, culm length, ear number and ear length by using 120 lines by anther culture, a cross between the Indica variety Cheongcheong and Japonica variety Nagdong. DNA marker was selected on the QTLs gene, and the following results were obtained. CNDH (Cheongcheong Nagdong Doubled Haploid) lines frequency distribution table curves about culm length, ear number and ear length exhibited showed a continuous variation close to a normal distribution. QTL analysis result, on culm length qPlL1-1 and qPlL1-2 were detected on the chromosome 1 and qPlL5 was detected on the chromosome 5. However, on ear length qPL2, qPL3 and qPL10, were detected on the chromosome 2, 3 and 10, while on ear number qPN1-1 and qPN1-2 were detected on the chromosome 1, qPN9 was detected on the chromosome 9. The QTLs related to culm length was found to chromosomes 5 and LOD scores were 3.81. The QTLs related to ear length was found to chromosomes 2 and 3 LOD scores were 7.13 and 3.20. The QTLs related to ear number was found to chromosome 9 and LOD scores were 4.27. Twenty two (22) Japonica cultivars and 12 Indica cultivars were analyzed polymorphisms, using selected 9 markers from the result about plant form analysis. RM5311, RM555 and RM8111 about the culm length, the ear length and number of ear were selected on the standard of Cheongcheong and Nagdong. Each rate of concordances about the culm length, the ear length and number of ear are 44.11%, 41.17% and 44.11%.

• Journal of the Korean earth science society
• /
• v.23 no.3
• /
• pp.280-293
• /
• 2002

In this study, we analyzed the long-term distribution patterns of $CH_4$ determined from the Moo-Ahn (MAN) observatory in relation with those derived from the world major background monitoring sites. Comparison of the data were made using those data sets collected for the period between Aug. 1995 to Dec. 1991. The mean $CH_4$ concentration of MAN observatory was measured to be 1898${\pm}$85.3 ppb, recording the highest concentration of all the monitoring sites. When the concentration of $CH_4$ for different stations was compared over latitudinal scale, its concentration appeared to increase systematically as a function of latitude with an exception of MAN (and the other Korean monitoring site at Tae Ahn). Moreover, such phenomenon was more distinctive in Northern than Southern Hemisphere. According to the analysis of the monthly distribution patterns of $CH_4$ at MAN observatory, its concentration level began to increase from the months of February/March and peaked during August. In addition, when the level of oscillation in monthly concentrations (between the maximum and minimum values) was checked, differences were significant between MAN and other monitoring stations. If the rate of concentration change was checked using the data sets collected for this limited time period in terms of linear regression analysis, results for MAN showed the highest annual increasing rate of 16.5 ppb. It is hence suggested that the largest variability in the $CH_4$ distribution patterns at MAN observatory may be reflected by the high irregularity in its source/sink processes.

• Journal of Venture Innovation
• /
• v.1 no.1
• /
• pp.49-65
• /
• 2018

For a long time the German economy was primarily defined by large corporations and thriving small and medium-sized enterprises. Since about 2005 a second strand has started to emerge and it is one which is becoming increasingly important and is creating jobs - start-ups in the digital sector. This start-up activity is taking an important role in Germany's economic development: Start-up companies spawn innovations and create jobs, thus promoting the concept of competition. In general "start-up" refers to digitally-driven companies that are not more than five years old. Germany's start-up policy consists of three main parts. First of all, Germany has the characteristics of technology-based start-ups. The Hartz reform since 2002 has shown its focus on technology-based start-ups. In particular, it is the most appropriate for a start-up company to take the role of a new technology company to respond to changes in the global industrial structure. Second, it is approaching from a long-term perspective. In this regard, the small business policy, including Germany's new business policy, is seen as a tradition that can be consistent and can make policy decisions based on the basics rather than following the times. Third, the government is implementing policies centered on demand. Germany's start-up policy is summarized as a technology-based policy and new job creation. The policy response is that the government seeks the best combination of policies by adapting them to the times from the broad trend of employment market policies. What is important here is that policies are made based on consumers, not suppliers, in the process of policy making and implementation. With the Digital Agenda 2020 the Federal government has likewise committed itself to preparing the digital economy for international competition and making Germany the "No. 1 digital growth country in Europe". Ever since 1998 the Federal Ministry for Economic Affairs and Energy (BMWi) has awarded the "EXIST" start-up scholarship to students and graduates. The Ministry also invests in the High Tech start-up fund. Together with Kreditanstalt für Wiederaufbau (KfW) and 18 other investors from the world of business the seed investor promotes young technology companies. Germany offers start-ups a good infrastructure and lots of funding opportunities. Berlin is regarded as Europe's start-up capital and also attracts lots of international young entrepreneurs.

• Journal of Technology Innovation
• /
• v.19 no.3
• /
• pp.1-29
• /
• 2011

Korean government built "2040 Science and Technology Future Vision" in order to show positive future scenarios and suggest a long-term guideline for a progress in science and technology. The S&T Future Vision was built based on an analysis of global megatrends and a prospect of domestic social change. After building S&T Future Vision, the "Government R&E Strategy"s was established as a follow-up action plan. The Government R&D Strategy consists of lists of future emerging technologies for future leadership, government R&D investment status and investment portfolio plans. Exploring future emerging technologies aggressively and making a governmental R&D strategic policy are requirements for national competitiveness, leadership in the world. Therefore search and selection for future emerging technologies is getting more and more important recently. Generally qualitative methodologies have been used such as expert-panel discussion method and portfolio analysis with expert valuation method in order to explore future technologies. These experts-based qualitative methodologies are well defined but lacking in some objectivity because size of expert-panels has limitations. We suggest a quantitative methodology, science mapping method to compensate this shortcoming in this study. There is another limitation related governmental R&D strategy which is that general R&D portfolios are static until a point of technology realization. We also propose a dynamic R&D investment portfolio which present different portfolios at a intermediate point and a point of technology realization. We expect this try with science mapping method and a dynamic R&D portfolio could strengthen strategic aspect of government R&D policy.

• International Commerce and Information Review
• /
• v.13 no.3
• /
• pp.3-29
• /
• 2011

The Internet has become a medium through which people engage in increasingly sophisticated transactions. Businesses and consumers now use the Internet to communicate and engage in commercial transactions creating a virtual worldwide marketplace. They fear that the determination of Internet jurisdiction could be uncertain because electronic commerce is not executed in one particular place. Until now, there are no specific rules in the model laws and conventions dealing with international jurisdiction in electronic commerce. Due to the fact that U.S. companies are at the forefront of Internet technology, litigation regarding electronic commerce in the U.S. is more advanced than anywhere else in the world This paper analysis the basic framework for personal jurisdiction and approach for determining international jurisdiction in electronic commerce cases and explain the differences of several approaches involving interactions over the Internet. According to jurisdiction approach test, the U.S. employs sliding scale, effects and targeting test in electronic commerce. In recent many research views the targeting test as a global standard for determining international electronic commerce jurisdiction. However, there is still no clear indication of conclusive test of jurisdiction determination for electronic commerce. Therefore, it is a changing and process of jurisdiction test in the U.S. cases. In Korea, there is jurisdiction related clause in Private International Law, but it may be asked whether applicable in electronic commerce. Accordingly, analysis of the precedents changing related to electronic commerce jurisdiction of U.s. is full of suggestions in Korean companies, consumers and helps an enactment of code of civil procedure that containing many group's demands.

• Journal of Environmental Policy
• /
• v.9 no.2
• /
• pp.157-184
• /
• 2010

Global climate change is destroying the water circulation balance by changing rates of precipitation, recharge and discharge, and evapotranspiration. The Intergovernmental Panel on Climate Change (IPCC 2007) makes "changes in rainfall pattern due to climate system changes and consequent shortage of available water resource" a high priority as the weakest part among the effects of human environment caused by future climate changes. Groundwater, which occupies a considerable portion of the world's water resources, is related to climate change via surface water such as rivers, lakes, and marshes, and "direct" interactions, being indirectly affected through recharge. Therefore, in order to quantify the effects of climate change on groundwater resources, it is necessary to not only predict the main variables of climate change but to also accurately predict the underground rainfall recharge quantity. In this paper, the authors selected a relevant climate change scenario, In this context, the authors selected A1B from the Special Report on Emission Scenario (SRES) which is distributed at Korea Meteorological Administration. By using data on temperature, rainfall, soil, and land use, the groundwater recharge rate for the research area was estimated by period and embodied as geographic information system (GIS). In order to calculate the groundwater recharge quantity, Visual HELP3 was used as main model for groundwater recharge, and the physical properties of weather, temperature, and soil layers were used as main input data. General changes to water circulation due to climate change have already been predicted. In order to systematically solve problems associated with how the groundwater resource circulation system should be reflected in future policies pertaining to groundwater resources, it may be urgent to recalculate the groundwater recharge quantity and consequent quantity for using via prediction of climate change in Korea in the future and then reflection of the results. The space-time calculation of changes to the groundwater recharge quantity in the study area may serve as a foundation to present additional measures for the improved management of domestic groundwater resources.

• Korea and Global Affairs
• /
• v.1 no.1
• /
• pp.71-96
• /
• 2017

This study speculates on responses to the nuclear threats of North Korea and mutual assistance and cooperation between South Korea, the U.S. and China for the unification of the Korean Peninsula. As for the North Koreas nuclear issue and unification of the Korean Peninsula, South Korea is the subject of national division, the U.S. is a responsible country in international issues and does not have diplomatic ties with North Korea. China is a traditional socialist nation and a supporter of North Korea. As North Korea's strategic weapons including nuclear weapons and ballistic missiles are international issues, to defend against Kim Jung-Eun's unexpected acts, the three countries should actively cooperate with each other and develop countermeasures. However, with respect to the road map of the North Koreas issue, there are subtle differences between the U.S. and China in recognition of and sanctions against North Korea as a resolution of the U..N. Security Council. The U.S. has continued a deterrence policy and sanctions against North Korea based on joint threats between South Korea and the U.S. while China has showed a negative position in the process of solving the North Korean nuclear issue because of the unstable security derived from the U.S. 's intervention in the Korean peninsula. North Korea should change its diplomatic policy in a more concrete way towards world peace although it has continued trade of strategic weapons with Middle Eastern countries to maintain its political system. For example, to restart the summit talks and open multilateral security channels. Although the issue of unification of the Korean peninsula should be resolved by South and North Korea themselves, it is strange that South and North Korea depend on the logic of powerful countries for the resolution of a national problem. As for North Koreas nuclear and the Unification issues, peaceful solutions presented by South Korea seem more persuasive than the solution presented by North Korea which did not secure any international support. However, South Korea, the U.S. and China need to develop uni-directional two-tract strategies for sanctions against North Korea and talks with North Korea for peace on the Korean peninsula, and should continue to support the economic independence of North Korea.