• 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.

• International Journal of Highway Engineering
• /
• v.8 no.4 s.30
• /
• pp.49-62
• /
• 2006

The republic of korea has put a great emphasis on the role of the road as widening a social infra-structural facility. Thus, vast amount of money has been invested on the road establishment. As a result, there has been fruitful outcomes in establishing the road system of the nation especially for the flat road with ease. However, in order to have more systematic and sustainable road system, we should turn our attention to more painful and high-cost regions such as mountainous districts and those are to be developed effectively. The configuration of the road is an important factor to be considered in making a decision for the road planning. Nevertheless, current road planning criterion has no such clarified and objective judging standard for figuring the configuration of the road out and, as a result, speed planning can be decided incorrectly. our research has acknowledged the necessity of estimating the configuration of the road and aimed to make it organized and sorted according to the height, slope, and the vehicle's speed. The results are as follows. First, our research made use of GIS data and classified the road into 9 different areas according to the height and the slope. Also, road classification being matched to the data of vehicle's speed, it has been shown that those characteristics of different areas have made an influence on vehicle's speed. Secondly, based on the results of the similarity between geographical classification and, vehicle's speed of sorted groups according to the height and the slope, conclusively we have classified as flat, rolling region and mountain. Since our research has made use of vehicle's speed for National Highway, it is not applicable to different functional highways. However, for the highway to be established hereafter, it can be a standard for reflection geographical characteristics.

• Journal of the Korean Chemical Society
• /
• v.54 no.5
• /
• pp.556-566
• /
• 2010

A new kinetic spectrophotometric method is developed for the measurement of Mn(II) in natural water samples. The method is based on the catalytic effect of Mn(II) with the oxidation of Gallocyanin by $KIO_4$ using nitrilotriacetic acid (NTA) as an activation reagent at 620 nm. The optimum conditions obtained are $4.00{\times}1^{-5}\;M$ Gallocyanin, $KIO_4$, $1.00{\times}10^{-4}\;M$ NTA, 0.1 M HAc/NaAc buffer of pH = 3.50, the reaction time of 5 min and the temperature of $30^{\circ}C$. Under the optimum conditions, the proposed method allows the measurement of Mn(II) in a range of $0.1\;-\;4.0\;ng\;mL^{-1}$ and with a detection limit of down to $0.025\;ng\;mL^{-1}$. The recovery efficiency in measuring the standard Mn(II) solution is in a range of 98.5 - 102%, and the RSD is in a range of 0.76 - 1.25%. The newly developed kinetic method has been successfully applied to the measurement of Mn(II) in both some environmental water samples and certified standard reference river water sample, JAC-0031 with satisfying results. Moreover, few cations and anions interfere with the measurement of Mn(II). Compared with the other catalytic-kinetic methods and instrumental methods, the proposed kinetic method shows fairly good selectivity and sensitivity, low cost, cheapness, low detection limit and rapidity. It can easily and successfully be applied to the real water samples with relatively low salt content and complex matrices such as bottled drinking water, cold and hot spring waters, lake water, river water samples.

• Journal of the Korean Wood Science and Technology
• /
• v.36 no.3
• /
• pp.30-38
• /
• 2008

Petroleum-based resin adhesives have extensively been used for the production of wood panels. However, with the increase of manufacturing cost and the environmental issue, such as the emission of volatile organic compounds, of the adhesive resins, it is necessary to be developed new adhesive systems. In this study, the potential of okara, which is a residue wasted from the production of tofu, for the development of bio-based adhesives was investigated. At first, the physical and chemical properties of okara were examined. After okara was hydrolyzed in acidic and/or alkaline solutions, okara-based adhesive resins were formulated with the mixtures of the okara hydrolyzates and phenol formaldehyde (PF) prepolymer. The adhesive resins were used for the fabrication of plywood panels, and then the adhesive strength and formaldehyde emission of the plywood panels were measured to examine the applicability of the resin adhesives for the production of plywood panels. The solids content and pH of the okara used in this study were around 20% and weak acidic state, respectively. In the analysis of its chemical composition, the content of carbohydrate was the highest, and followed by protein. The shear strengths of plywood fabricated with okara-based resin adhesives exceeded a minimum requirement of KS standard for ordinary plywood, but its wood failure did not reach the minimum requirement. In addition, the formaldehyde emissions of all plywood panels were higher than that of E1 specified in the KS standard. Based on these results, okara has the potential to be used as a raw material of environmentally friendly adhesive resin systems for the production of wood panels, but further researches - biological hydrolysis of okara and various formulations of PF prepolymer - are required to improve the adhesive strength and formaldehyde emission of okara-based resin adhesives.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.49 no.2
• /
• pp.27-40
• /
• 2021

This study was carried out for the description of the conflicts on the measurement of the root collar diameter of the landscape trees that are currently being produced, distributed, and planted in S. Korea, and for determination of the standard for root collar diameter measurement. The difference in consciousness of appropriate measurement of root collar diameter among different ages of landscape practitioners was statistically significant at p<0.05 level. It seemed to be due to the difference in the amount of field experiences among different age groups. On "the ambiguity of measuring the root collar diameter' of landscape trees", the consciousness was significantly different at p<0.05 level among job positions. On "Improvement of measurement criteria for landscape trees," it was significantly different at p<0.05 level among job types. This was thought to be due to the disagreement between the client and the contractor. On "prevention of topsoil removal" when excavating landscape trees, the consciousness was significantly different at p<0.001 level among different age groups, and different at p<0.01 level among different occupations, and different at p<0.05 level among different working area. The consciousness on "removing top soil when excavating landscape trees and rooting after transplantation" was not significantly different. The consciousness on the conflict caused by "ambiguity in root collar diameter measurement criteria" was high with an average of 3.85 for job type, occupation, jop position, and work experience. It was higher for landscape contractors than public institutions. The higher job positions and more experiences, the more conflicts. The consciousness on the appropriate position of root collar diameter measurement for landscape trees revealed that measuring at above-ground part (66.5%) was prefered to the underground part (33.0%). During the excavation of landscape trees for transplant, topsoil removal up to average depth of -2cm to -4cm was favored by 84.0%, and the purpose of removing topsoil was recognized as 'to increase the size and unit cost' by 59.7%.

• Korean Journal of Remote Sensing
• /
• v.39 no.5_1
• /
• pp.669-681
• /
• 2023

Studies have tried to apply remote sensing technology, a non-destructive survey method, instead of the existing destructive survey, which requires relatively large labor input and a long time to estimate chlorophyll content, which is an important indicator for evaluating the growth of fruit trees. This study was conducted to non-destructively evaluate the chlorophyll content of pear tree leaves using unmanned aerial vehicle-based hyperspectral imagery for two years(2021, 2022). The reflectance of the single bands of the pear tree canopy extracted through image processing was band rationed to minimize unstable radiation effects depending on time changes. The estimation (calibration and validation) models were developed using machine learning algorithms of elastic-net, k-nearest neighbors(KNN), and support vector machine with band ratios as input variables. By comparing the performance of estimation models based on full band ratios, key band ratios that are advantageous for reducing computational costs and improving reproducibility were selected. As a result, for all machine learning models, when calibration of coefficient of determination (R²)≥0.67, root mean squared error (RMSE)≤1.22 ㎍/cm², relative error (RE)≤17.9% and validation of R²≥0.56, RMSE≤1.41 ㎍/cm², RE≤20.7% using full band ratios were compared, four key band ratios were selected. There was relatively no significant difference in validation performance between machine learning models. Therefore, the KNN model with the highest calibration performance was used as the standard, and its key band ratios were 710/714, 718/722, 754/758, and 758/762 nm. The performance of calibration showed R²=0.80, RMSE=0.94 ㎍/cm², RE=13.9%, and validation showed R²=0.57, RMSE=1.40 ㎍/cm², RE=20.5%. Although the performance results based on validation were not sufficient to estimate the chlorophyll content of pear tree leaves, it is meaningful that key band ratios were selected as a standard for future research. To improve estimation performance, it is necessary to continuously secure additional datasets and improve the estimation model by reproducing it in actual orchards. In future research, it is necessary to continuously secure additional datasets to improve estimation performance, verify the reliability of the selected key band ratios, and upgrade the estimation model to be reproducible in actual orchards.

• Nuclear Medicine and Molecular Imaging
• /
• v.43 no.4
• /
• pp.323-329
• /
• 2009

Purpose: Radioimmunoassay (RIA) was usually performed by the batch assay. To improve the efficiency of RIA without increase of the cost and time, random assay could be a choice. We investigated the possibility of the random assay using automatic dispenser by assessing the agreement between batch assay and random assay. Materials and Methods: The experiments were performed with four items; Triiodothyronine (T3), free thyroxine (fT4), Prostate specific antigen (PSA), Carcinoembryonic antigen (CEA). In each item, the sera of twenty patients, the standard, and the control samples were used. The measurements were done 4 times with 3 hour time intervals by random assay and batch assay. The coefficient of variation (CV) of the standard samples and patients' data in T3, fT4, PSA, and CEA were assessed. ICC (Intraclass correlation coefficient) and coefficient of correlation were measured to assessing the agreement between two methods. Results: The CVs (%) of T3, fT4, PSA, and CEA measured by batch assay were 3.2$\pm$1.7%, 3.9$\pm$2.1%, 7.1$\pm$6.2%, 11.2$\pm$7.2%. The CVs by random assay were 2.1$\pm$1.7%, 4.8$\pm$3.1%, 3.6$\pm$4.8%, and 7.4$\pm$6.2%. The ICC between the batch assay and random assay were 0.9968 (T3), 0.9973 (fT4), 0.9996 (PSA), and 0.9901 (CEA). The coefficient of correlation between the batch assay and random assay were 0.9924(T3), 0.9974 (fT4), 0.9994 (PSA), and 0.9989 (CEA) (p<0.05). Conclusion: The results of random assay showed strong agreement with the batch assay in a day. These results suggest that random assay using automatic dispenser could be used in radioimmunoassay.

• Journal of Korean Home Economics Education Association
• /
• v.30 no.4
• /
• pp.141-164
• /
• 2018

The purpose of this study was to investigate the stress perception, dietary habits, lifestyle and snack intake behaviors of students to help dietary education of high school students in Jeonnam area. The data were collected from high school students in Jeonnam area and 503 questionnaires were used for analysis. The main results of this study are as follows. First, the mean scores of high school students' perception of stress were 2.92 points for male students and 3.07 points for female students, and female students were statistically significantly higher than male students(p<.01). Stress perceptions showed significant differences in school stress(p<.001) and family stress(p<.05) according to gender, and female students gained more stressed than male students. Second, the dietary habits and lifestyle characteristics of high school students by gender showed significant difference between male students and female students at the one meal time(p<.001), the amount of meal(p<.05), the tendency of eating habits(p<.001), the salty taste of food(p<.05), the exercise frequency per week(p<.001) and the exercise time per day(p<.001), respectively. One meal time was shorter for male students than for female students, and the amount of meals was more suitable amount for female students than male students. Also, male students ate more regularly than female students and female students preferred more stimulating foods than male students. Male students ate slightly bland salty taste, and female students ate standard salty taste. The exercise frequency per week and the exercise time per day were more in male students than in female students. Third, the snack intake characteristics of high school students by gender showed significantly different between male students and female students at the cost of snack per day(p<.01), the time to eat snacks(p<.01) and the favorite snacks(p<.001), respectively. The cost of snack per day was higher for female students than for male students in case of less than 4,000 won. The time for eating snack showed that 'Irregularly' was high for male students and 'In the evening after lunch' was high for female students. The favorite snacks were cookies, candies and chocolates for male and female students. Forth, the dietary habits and lifestyle characteristics of high school students by stress level showed significantly different at the frequency of daily meals(p<.05), the tendency of eating habits(p<.01) and the salty taste of food(p<.05), respectively. The frequency of daily meals was more than four times in case of the 'High stress' and 'Low stress' groups. The tendency of eating habit was low in regular meal rate in 'High stress' group. Fifth, the snack intake characteristics of high school students by stress level showed significantly different in the frequency of snacks per day(p<.05) and the favorite taste of snacks(p<.05). In case of the frequency of daily snack intake, 'More than 4 times' was highest in 'High stress' group and 'Low stress' group. The favorite taste of snacks showed that the spicy taste was highest in the 'High stress' group and the 'Low stress' group.

• Journal of Intelligence and Information Systems
• /
• v.20 no.3
• /
• pp.109-131
• /
• 2014

As the demand of nuclear power plant equipment is continuously growing worldwide, the importance of handling nuclear strategic materials is also increasing. While the number of cases submitted for the exports of nuclear-power commodity and technology is dramatically increasing, preadjudication (or prescreening to be simple) of strategic materials has been done so far by experts of a long-time experience and extensive field knowledge. However, there is severe shortage of experts in this domain, not to mention that it takes a long time to develop an expert. Because human experts must manually evaluate all the documents submitted for export permission, the current practice of nuclear material export is neither time-efficient nor cost-effective. Toward alleviating the problem of relying on costly human experts only, our research proposes a new system designed to help field experts make their decisions more effectively and efficiently. The proposed system is built upon case-based reasoning, which in essence extracts key features from the existing cases, compares the features with the features of a new case, and derives a solution for the new case by referencing similar cases and their solutions. Our research proposes a framework of case-based reasoning system, designs a case-based reasoning system for the control of nuclear material exports, and evaluates the performance of alternative keyword extraction methods (full automatic, full manual, and semi-automatic). A keyword extraction method is an essential component of the case-based reasoning system as it is used to extract key features of the cases. The full automatic method was conducted using TF-IDF, which is a widely used de facto standard method for representative keyword extraction in text mining. TF (Term Frequency) is based on the frequency count of the term within a document, showing how important the term is within a document while IDF (Inverted Document Frequency) is based on the infrequency of the term within a document set, showing how uniquely the term represents the document. The results show that the semi-automatic approach, which is based on the collaboration of machine and human, is the most effective solution regardless of whether the human is a field expert or a student who majors in nuclear engineering. Moreover, we propose a new approach of computing nuclear document similarity along with a new framework of document analysis. The proposed algorithm of nuclear document similarity considers both document-to-document similarity (${\alpha}$) and document-to-nuclear system similarity (${\beta}$), in order to derive the final score (${\gamma}$) for the decision of whether the presented case is of strategic material or not. The final score (${\gamma}$) represents a document similarity between the past cases and the new case. The score is induced by not only exploiting conventional TF-IDF, but utilizing a nuclear system similarity score, which takes the context of nuclear system domain into account. Finally, the system retrieves top-3 documents stored in the case base that are considered as the most similar cases with regard to the new case, and provides them with the degree of credibility. With this final score and the credibility score, it becomes easier for a user to see which documents in the case base are more worthy of looking up so that the user can make a proper decision with relatively lower cost. The evaluation of the system has been conducted by developing a prototype and testing with field data. The system workflows and outcomes have been verified by the field experts. This research is expected to contribute the growth of knowledge service industry by proposing a new system that can effectively reduce the burden of relying on costly human experts for the export control of nuclear materials and that can be considered as a meaningful example of knowledge service application.

• Korean Journal of Weed Science
• /
• v.13 no.4
• /
• pp.210-233
• /
• 1993

The herbicide has become indispensable as much as nitrogen fertilizer in Korean agriculture from 1970 onwards. It is estimated that in 1991 more than 40 herbicides were registered for rice crop and treated to an area 1.41 times the rice acreage ; more than 30 herbicides were registered for field crops and treated to 89% of the crop area ; the treatment acreage of 3 non-selective foliar-applied herbicides reached 2,555 thousand hectares. During the last 25 years herbicides have benefited the Korean farmers substantially in labor, cost and time of farming. Any herbicide which causes crop injury in ordinary uses is not allowed to register in most country. Herbicides, however, can cause crop injury more or less when they are misused, abused or used under adverse environments. The herbicide use more than 100% of crop acreage means an increased probability of which herbicides are used wrong or under adverse situation. This is true as evidenced by that about 25% of farmers have experienced the herbicide caused crop injury more than once during last 10 years on authors' nationwide surveys in 1992 and 1993 ; one-half of the injury incidences were with crop yield loss greater than 10%. Crop injury caused by herbicide had not occurred to a serious extent in the 1960s when the herbicides fewer than 5 were used by farmers to the field less than 12% of total acreage. Farmers ascribed about 53% of the herbicidal injury incidences at their fields to their misuses such as overdose, careless or improper application, off-time application or wrong choice of the herbicide, etc. While 47% of the incidences were mainly due to adverse natural conditions. Such misuses can be reduced to a minimum through enhanced education/extension services for right uses and, although undesirable, increased farmers' experiences of phytotoxicity. The most difficult primary problem arises from lack of countermeasures for farmers to cope with various adverse environmental conditions. At present almost all the herbicides have"Do not use!" instructions on label to avoid crop injury under adverse environments. These "Do not use!" situations Include sandy, highly percolating, or infertile soils, cool water gushing paddy, poorly draining paddy, terraced paddy, too wet or dry soils, days of abnormally cool or high air temperature, etc. Meanwhile, the cultivated lands are under poor conditions : the average organic matter content ranges 2.5 to 2.8% in paddy soil and 2.0 to 2.6% in upland soil ; the canon exchange capacity ranges 8 to 12 m.e. ; approximately 43% of paddy and 56% of upland are of sandy to sandy gravel soil ; only 42% of paddy and 16% of upland fields are on flat land. The present situation would mean that about 40 to 50% of soil applied herbicides are used on the field where the label instructs "Do not use!". Yet no positive effort has been made for 25 years long by government or companies to develop countermeasures. It is a really sophisticated social problem. In the 1960s and 1970s a subside program to incoporate hillside red clayish soil into sandy paddy as well as campaign for increased application of compost to the field had been operating. Yet majority of the sandy soils remains sandy and the program and campaign had been stopped. With regard to this sandy soil problem the authors have developed a method of "split application of a herbicide onto sandy soil field". A model case study has been carried out with success and is introduced with key procedure in this paper. Climate is variable in its nature. Among the climatic components sudden fall or rise in temperature is hardly avoidable for a crop plant. Our spring air temperature fluctuates so much ; for example, the daily mean air temperature of Inchon city varied from 6.31 to $16.81^{\circ}C$ on April 20, early seeding time of crops, within${\times}$2Sd range of 30 year records. Seeding early in season means an increased liability to phytotoxicity, and this will be more evident in direct water-seeding of rice. About 20% of farmers depend on the cold underground-water pumped for rice irrigation. If the well is deep over 70m, the fresh water may be about $10^{\circ}C$ cold. The water should be warmed to about $20^{\circ}C$ before irrigation. This is not so practiced well by farmers. In addition to the forementioned adverse conditions there exist many other aspects to be amended. Among them the worst for liquid spray type herbicides is almost total lacking in proper knowledge of nozzle types and concern with even spray by the administrative, rural extension officers, company and farmers. Even not available in the market are the nozzles and sprayers appropriate for herbicides spray. Most people perceive all the pesticide sprayers same and concern much with the speed and easiness of spray, not with correct spray. There exist many points to be improved to minimize herbicidal phytotoxicity in Korea and many ways to achieve the goal. First of all it is suggested that 1) the present evaluation of a new herbicide at standard and double doses in registration trials is to be an evaluation for standard, double and triple doses to exploit the response slope in making decision for approval and recommendation of different dose for different situation on label, 2) the government is to recognize the facts and nature of the present problem to correct the present misperceptions and to develop an appropriate national program for improvement of soil conditions, spray equipment, extention manpower and services, 3) the researchers are to enhance researches on the countermeasures and 4) the herbicide makers/dealers are to correct their misperceptions and policy for sales, to develop database on the detailed use conditions of consumer one by one and to serve the consumers with direct counsel based on the database.