• Asia pacific journal of information systems
• /
• v.19 no.2
• /
• pp.157-178
• /
• 2009

Corporate credit rating is a very important factor in the market for corporate debt. Information concerning corporate operations is often disseminated to market participants through the changes in credit ratings that are published by professional rating agencies, such as Standard and Poor's (S&P) and Moody's Investor Service. Since these agencies generally require a large fee for the service, and the periodically provided ratings sometimes do not reflect the default risk of the company at the time, it may be advantageous for bond-market participants to be able to classify credit ratings before the agencies actually publish them. As a result, it is very important for companies (especially, financial companies) to develop a proper model of credit rating. From a technical perspective, the credit rating constitutes a typical, multiclass, classification problem because rating agencies generally have ten or more categories of ratings. For example, S&P's ratings range from AAA for the highest-quality bonds to D for the lowest-quality bonds. The professional rating agencies emphasize the importance of analysts' subjective judgments in the determination of credit ratings. However, in practice, a mathematical model that uses the financial variables of companies plays an important role in determining credit ratings, since it is convenient to apply and cost efficient. These financial variables include the ratios that represent a company's leverage status, liquidity status, and profitability status. Several statistical and artificial intelligence (AI) techniques have been applied as tools for predicting credit ratings. Among them, artificial neural networks are most prevalent in the area of finance because of their broad applicability to many business problems and their preeminent ability to adapt. However, artificial neural networks also have many defects, including the difficulty in determining the values of the control parameters and the number of processing elements in the layer as well as the risk of over-fitting. Of late, because of their robustness and high accuracy, support vector machines (SVMs) have become popular as a solution for problems with generating accurate prediction. An SVM's solution may be globally optimal because SVMs seek to minimize structural risk. On the other hand, artificial neural network models may tend to find locally optimal solutions because they seek to minimize empirical risk. In addition, no parameters need to be tuned in SVMs, barring the upper bound for non-separable cases in linear SVMs. Since SVMs were originally devised for binary classification, however they are not intrinsically geared for multiclass classifications as in credit ratings. Thus, researchers have tried to extend the original SVM to multiclass classification. Hitherto, a variety of techniques to extend standard SVMs to multiclass SVMs (MSVMs) has been proposed in the literature Only a few types of MSVM are, however, tested using prior studies that apply MSVMs to credit ratings studies. In this study, we examined six different techniques of MSVMs: (1) One-Against-One, (2) One-Against-AIL (3) DAGSVM, (4) ECOC, (5) Method of Weston and Watkins, and (6) Method of Crammer and Singer. In addition, we examined the prediction accuracy of some modified version of conventional MSVM techniques. To find the most appropriate technique of MSVMs for corporate bond rating, we applied all the techniques of MSVMs to a real-world case of credit rating in Korea. The best application is in corporate bond rating, which is the most frequently studied area of credit rating for specific debt issues or other financial obligations. For our study the research data were collected from National Information and Credit Evaluation, Inc., a major bond-rating company in Korea. The data set is comprised of the bond-ratings for the year 2002 and various financial variables for 1,295 companies from the manufacturing industry in Korea. We compared the results of these techniques with one another, and with those of traditional methods for credit ratings, such as multiple discriminant analysis (MDA), multinomial logistic regression (MLOGIT), and artificial neural networks (ANNs). As a result, we found that DAGSVM with an ordered list was the best approach for the prediction of bond rating. In addition, we found that the modified version of ECOC approach can yield higher prediction accuracy for the cases showing clear patterns.

• Korean Journal of Ecology and Environment
• /
• v.51 no.1
• /
• pp.71-85
• /
• 2018

A recent topic of interest within ecology is ecosystem services that are importantly connected with biodiversity. Biodiversity is a fundamental component of ecological studies and has been the subject of many studies worldwide. However, studies on the diversity and distribution of freshwater fish were not undertaken in earnest until 2000, and the number of internationally available studies is therefore still limited. In this study, we used data from "The Survey and Evaluation of Aquatic Ecosystem Health (hereafter SEAEH)" within the framework of which whole water systems on the Korean peninsula are annually monitored to investigate and determine the current status of the diversity and distribution of freshwater fish, as well as to elucidate major factors impacting freshwater fish. A total of 130 species from 28 families were identified at around 953 sites in the SEAEH investigation of 2014. The species Zacco platypus (relative abundance: 28.2%) and Z. koreanus (19.3%) were identified as the most abundant species on the Korean peninsula, and besides these 20 endangered species, 51 endemic species and 4 exotic species were also collected. Highest (96 species) and lowest (72 species) diversity were identified in the Han River Watershed and the Yeongsan/Seomjin River Watershed, respectively. The mean endemic rate was 32.7%, which is higher than the previous rate determined in 1995. The exotic species Micropterus salmoides, Lepomis macrochirus and Carassius cuvieri were found in the whole river system except north Yeongdong, whereas Oreochromis niloticus occurred only at three sites due to the cold water temperatures in winter. Totally, 28 species were found to be translocated to different water systems not connected with known original habitats, most translocations occurring in the Nakdong River and the Yeongdong region. Among the translocated species, Opsariichthys uncirostris amurensis, and Erythroculter erythropterus were dominantly distributed in the mid- to lower reaches of Nakdong River where they may be harmful to the native fish assemblages due to their strong predation. The construction of weirs in connection with the "4 River Project" generated changes in fish assemblages before, during and after the project, resulting in altered habitat condition. Especially, a decrease in the number of endemic species and an increase in the abundance of lentic species were seen. Human-induced global warming will allow exotic species requiring high water temperatures to adapt to the environmental conditions of the Korean peninsula, which may enhance the diversity of native fishes and create changes in distribution. For the establishment of realistic and efficient management and protection strategies for Korean freshwater fish, SEAEH, which produces fundamental and quantitative data for Korean streams, is an important and necessary tool.

• Journal of Life Science
• /
• v.27 no.10
• /
• pp.1137-1144
• /
• 2017

The microalgae Chlorella vulgaris has been considered an important alternative resource for biodiesel production. However, its industrial-scale production has been constrained by the low productivity of the biomass and lipid. To overcome this problem, we isolated and characterized a potentially economical oleaginous strain of C. vulgaris via the random mutagenesis technique using UV irradiation. Two types of mass production systems were compared for their yield of biomass and lipid content. Among the several putatively oleaginous strains that were isolated, the particular mutant strain designated as UBM1-10 in the laboratory showed an approximately 1.5-fold higher cell yield and lipid content than those from the wild type. Based on these results, UBM1-10 was selected and cultivated under outdoor conditions using two different types of reactors, a tubular-type photobioreactor (TBPR) and an open pond-type reactor (OPR). The results indicated that the mutant strain cultivated in the TBPR showed more than 5 times higher cell concentrations ($2.6g\;l^{-1}$) as compared to that from the strain cultured in the OPR ($0.5g\;l^{-1}$). After the mass cultivation, the cells of UBM1-10 and the parental strain were further investigated for crude lipid content and composition. The results indicate a 3-fold higher crude lipid content from UBM1-10 (0.3%, w/w) as compared to that from the parent strain (0.1% w/w). Therefore, this study demonstrated that the economic potential of C. vulgaris as a biodiesel production resource can be increased with the use of a photoreactor type as well as the strategic mutant isolation technique.

• Korean Journal of Food Science and Technology
• /
• v.35 no.1
• /
• pp.103-110
• /
• 2003

Stable production of fermented kanjang containing 1.8% (v/v) ethanol was obtained within four days using traditional kanjang containing 4% added glucose in packed-bed bioreactor systems filled with immobilized Zygosaccharomyces rouxii and Candida versatilis on porous alumina ceramic bead carrier at $28{\pm}0.5^{\circ}C$ and aeration rate of 0.05 vvm. Specific rates of alcohol production for Z. rouxii and C. versatilis were 0.0033 and 0.0031/day, respectively, and those of glucose consumption were both -0.0087/day in the batch type of alcoholic fermentation. In semi-continuous alcoholic fermentation at a dilution rate of 0.25/day, specific rates of alcohol production for Z. rouxii and C. versatilis were 0.0045 and 0.0029/day, and those of glucose consumption were -0.01 and -0.008/day, respectively, using identical bioreactor system. Similar specific rates of alcohol production were observed both in the batch or semi-continuous process and in the continuous one at the dilution rate of 0.25/day. Sensory characteristics of all alcoholic-fermented kanjang by Z. rouxii, C. versatilis, and a mixture of both yeasts (2:1, w/w) were shown to be significantly superior to those of home-made kanjang as revealed through organoleptic evaluation tests (p<0.05).

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.8 no.3
• /
• pp.52-63
• /
• 2009

Recently in Korea, in the case of metropolis, the urban signalized intersections are controlled by traffic information center or ITS center. Cheon-an City also established traffic information center through the 1st.-$\sim$3rd. ITS public construction and has managed this center that includes bus information service, traffic information collection and providing service, parking information service, and traffic responsive control system. In the Cheon-an metropolitan traffic signal operation, traffic signal controllers were grouped by the each main traffic flow axes and performed with coordinated signal timing for the signalized arterials, and also cycle and split changed by realtime traffic demands. Cheon-an urban traffic responsive control system was evaluated by intersection delay and speed, then it was verified that the delay decreased and vehicle speed improved. However, the rural signal control system to connect adjacency town was evaluated to have lower status than urban area due to the unimproved TOD (Time of day) plan. Therefore actuated signal control was examined for substitutive control system in isolated signal intersection. The aim of this article is to compare actuated signal control with TOD mode in the rural intersection of Cheon-an and to fine superiority of these two control mode, with evaluation of vehicle delay by using HCM(2000) method and by micro-simulation CORSlM. The result of field test show that actuated signal control gave better performance in delay comparison than the existing TOD signal control. And simulation outcome verified that non-optimized TOD has higher delay than optimized TOD mode, non-optimal actuated mode, and optimal actuated signal control mode. Particularly, these three modes delays had not different values according to the paired sample t-test. This is because small traffic demands were loaded in each links. This suggested actuated signal control is expected to be more effective than TOD mode in some rural isolated intersections which frequently need to survey for traffic volume.

• The Journal of Information Systems
• /
• v.25 no.1
• /
• pp.1-23
• /
• 2016

Purpose Self-congruity deals with the effect of symbolic value-expressive attributes on consumer decision and behavior, which is the theoretical foundation of the "non-utilitarian destination positioning". Functional congruity refers to utilitarian evaluation of a product or service by consumers. In addition, recent years, social network services, especially mobile social network services have created many opportunities for e-WOM communication that enables consumers to share personal consumption related information anywhere at any time. Moreover, self-construal is a hot and popular topic that has been discussed in the field of modem psychology as well as in marketing area. This study aims to examine the moderating effect of self-construal on the relationship between self-congruity, functional congruity and tourists' positive electronic word of mouth (e-WOM). Design/methodology/approach In order to verify the hypotheses, we developed a questionnaire with 32 survey items. We measured all the items on a five-point Likert-type scale. We used Sojump.com to collect questionnaire and gathered 218 responses from whom have visited Korea before. After a pilot test, we analyzed the main survey data by using SPSS 20.0 and AMOS 18.0, and employed structural equation modeling to test the hypotheses. We first estimated the measurement model for its overall fit, reliability and validity through a confirmatory factor analysis and used common method bias test to make sure that whether measures are affected by common-method variance. Then we tested the hypotheses through the structural model and used regression analysis to measure moderating effect of self-construal. Findings The results reveal that the effect of self-congruity on tourists' positive e-WOM is stronger for tourists with an independent self-construal compared with those with interdependent self-construal. Moreover, it shows that the effect of functional congruity on tourists' positive e-WOM becomes salient when tourists' self-construal is primed to be interdependent rather than independent. We expect that the results of this study can provide important implications for academic and practical perspective.

• Management & Information Systems Review
• /
• v.36 no.5
• /
• pp.45-60
• /
• 2017

This study aims to elucidate the key improvements through the current state of customer support for smart home services based on the Internet of things and the evaluation of user's usage. Smart home services typically provide a wide range of value in terms of security, safety, manageability (electricity and water use), convenience, and remote management accessibility. In this study, we analyzed the current state of smart home service based on Internet of Samsung, SKT and LG U + companies in Korea. However, since LG U+ is the only company providing user reviews, there is a limit to generalization, but we are trying to figure out whether the customer value is conveyed properly or not, and in which part the customer support is focused to support the service. As a result of analyzing the results of the study, we found that the smart home service is commercialized and marketed in various forms. However, it is questionable whether the technological level and user satisfaction level are sufficiently satisfied. The results of this study are as follows. First, although each company provides usage guidance, they still ask many questions about joining products and using products. Second, there are many defects in the product itself, and it is found that the companies are not satisfied with the overall response. Third, the three companies are focusing on switches, outlets, sensors, and lamps. This is an individual intelligent product rather than an interlocking or linking level, and it can be seen that there are many parts that are not compatible with the concept of the original Internet of things. In conclusion, this study shows that there are still many areas to improve on the level of customer service provision of smart home service, in particular, the ease of use is low and the quality of products is not reliable. We would like to present the improvement of this in detail through this study and reflect the companies that provide it and the service providers.

• Journal of Korea Water Resources Association
• /
• v.41 no.3
• /
• pp.277-291
• /
• 2008

In order to carry out river basin management, it is necessary to evaluate the state of the river basin and make site-specific measures on the basis of management goals and objectives. A river basin is divided into several watersheds, which are composed of several components: water resources, social and economic systems, law and institution, user, land, ecosystems, etc. They are connected among them and form network holistically. In this study, a methodology for evaluating watershed management was developed by consideration of the various features of a watershed system. This methodology employed factor analysis to develop sub-indexes for evaluating water use management, environment and ecosystem management, and flood management in a watershed. To do this, first, the related data were gathered and classified into six groups that are the components of watershed systems. Second, in all sub-indexes, preliminary tests such as KMO (Kaiser-Meyer-Olkin) measure of sampling adequacy and Bartlett's test of sphericity were conducted to check the data's acceptability to factor analysis, respectively. Third, variables related to each sub-index were grouped into three factors by consideration of statistic characteristics, respectively. These factors became indicators and were named, taking into account the relationship and the characteristics of included variables. In order to check the study results, the computed factor loadings of each variable were reviewed, and correlation analysis among factor scores was fulfilled. It was revealed that each factor score of factors in a sub-index was not correlated, and grouping variables by factor analysis was appropriate. And, it was thought that this indicator system would be applied effectively to evaluating the states of watershed management.

• The Korean Journal of Nuclear Medicine
• /
• v.38 no.6
• /
• pp.528-531
• /
• 2004

Purpose: PET has some disadvantage in the imaging of small animal due to poor resolution. With the advent of microPET scanner, it is possible to image small animals. However, the image quality was not good enough as human image. Due to larger brain, cat brain imaging was superior to mouse or rat. In this study, we established the cat brain infarction model and evaluate it and its temporal charge using microPET scanner. Materials and Methods: Two adult male cats were used. Anesthesia was done with xylazine and ketamine HCl. A burr hole was made at 1cm right lateral to the bregma. Collagenase type IV 10 ${\mu}l$ was injected using 30 G needle for 5 minutes to establish the infarction model. $^{18}F$-FDG microPET (Concorde Microsystems Inc., Knoxville, TN) scans were performed 1, 11 and 32 days after the infarction. In addition, $^{18}F$-FDG PET scans were performed using human PET scanner (Gemini, Philips medical systems, CA, USA) 13 and 47 days after the infarction. Results: Two cat brain infarction models were established. The glucose metabolism of an infarction lesion improved with time. An infarction lesion was also distinguishable in the human PET scan. Conclusion: We successfully established the cat brain infarction model and evaluated the infarcted lesion and its temporal change using $^{18}F$-FDG microPET scanner.

• Korean Journal of Environmental Agriculture
• /
• v.30 no.3
• /
• pp.310-315
• /
• 2011

BACKGROUND: Heavy-metal pollution represents an important environmental problem due to the toxic effects of metals, and their accumulation throughout the food chain leads to serious ecological and health problems. METHODS AND RESULTS: Optimum conditions in continuous-flow stirred tank reactor (CSTR) and packedbed column contactor (PBCC) using brown seaweed biosorbent were investigated. Under optimum conditions from both lab-scale biosorbent systems, removal efficiency of copper (Cu) in a large-scale PBCC system was investigated. Removal capacity of Cu using brown seaweed biosorbent in a lab-scale CSTR system was higher than that in a lab-scale PBCC system. On the other hand, over 48 L/day of flow rate in Cu solution, removal efficiency of Cu in a lab-scale PBCC system was higher than that in a lab-scale CSTR system. Optimum flow rate of Cu was 24 L/day, optimum Cu solution concentration was 100 mg/L. Removal capacity of Cu at different stages was higher in the order of double column biosorption system > single column biosorption system. Under different heavy metals, removal capacities of heavy metal were higher in the order of Pb > Cr > Ni > Mn ${\geq}$ Cu ${\geq}$ Cd ${\fallingdotseq}$ Zn ${\geq}$ Co. Removal capacity of Cu was 138 L in a large-scale PBCC system. Removal capacity of Cu a large-scale PBCC system was similar with in a lab-scale PBCC system. CONCLUSION(s): Therefore, PBCC system using brown seaweed biosorbent was suitable for treating heavy metal wastewater.