• 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.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.16 no.10
• /
• pp.6852-6859
• /
• 2015

The purpose of this study was to determine the growth characteristics and distribution pattern of a brackish water clam Corbicula japonica in Seomjin River. Field samples were taken from 14 stations with salinity gradients during spring. Salinity at the bottom layer ranged from 1.0 psu to 32.9 psu, with low salinities in the upper area of the river. In particular, salinity at St.11 was decreased drastically to be ca. 15.0 psu, indicating an intermediate salinity zone. The distribution pattern of C. japonica was related to the salinity gradient, with the highest densities of $2,102ind.m^{-2}$ at Station 13, followed by $1,507ind.m^{-2}$ at Station 11. Here, we focused on the growth characteristics of collected C. japonica collected at two stations with different salinity values. The relationship between shell length and total weight was highly correlated ($R^2=0.91$, P<0.001) at Station 13 compared to that at Station 11 ($R^2=0.72$, P<0.001). On the other hands, the degree of correlation between shell length and shell height (SH) or shell width (SW) at Station 11 (SH: $R^2=0.91$, P<0.001; SW: $R^2=0.69$, P<0.001) was higher than that at Station 13 (SH: $R^2=0.64$, P<0.001; SW: $R^2=0.48$, P<0.001). In addition, fatness index of C. japonica at Station 13 was significantly (P < 0.001) higher than that at St. 11 (t-test value=-22.8, p<0.001). This implies that C. japonica at Station 13 might have enhanced their somatic growth, whereas C. japonica at Station 11 might have this kind of defense mechanism their internal organization against the salinity stress. Ecologically, this kind of defense mechanism of C. japonica against salinity flucuation may play an important role in their survival strategy.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.2 no.2
• /
• pp.105-133
• /
• 1969

The stability of three major pigments in lavers, namely chlorophylls, carotenoids, and phy-cobilins, during processing and storage was studied. The results of the pigment retention in fresh lavers during storage at low temperatures, of heat lability during heat treatment, and of the pigment loss during storage are summarized as follows: 1. The contents of chlorophyll a, xanthophyll (lutein+zeaxanthin), and carotene ($\alpha-+\beta-carotene$) averaged 1,525, 627, and $409{\mu}g/g$ of dry samples respectively The comparative composition of four carotenoids, lutein, $\beta-carotene$, zeaxanthin, and $\alpha-carotene$, was 35.9, 33.7, 12.2, and $5.8\%$ respectively. 2. In storage of fresh lavers at room ternperature ($20-25^{\circ}C$), 2 to $5^{\circ}C$, and $-15^{\circ}C$, the pigments generally were retained more at 2 to $5^{\circ}C$ than others while carotene only showed more retention at $-15^{\circ}C$ than at 2 to $5^{\circ}C$. Phycocyanin tended to decrease more rapidly than phyco-erythrin at low temperatures. 3. In regard to heat treatment of both dried and wet materials at 40, 60, 80, $100^{\circ}C$ respectively, the pigments were more stable at lower temperatures than at higher temperatures in both cases of materials. Carotenoids were retained more in wet materials than in the sun dried. Xanthophylls showed higher heat stability than carotenes. Phycoerythrin was considered more heat labile than phycocyanin, particularly at higher temperatures. 4. In storage of heat treated lavers for 50 days at room temperature, the retention of pigments appeared differently in accordance with the temperatures of treatment The materials treated at $60^{\circ}C$ showed better relention of pigments than those treated at the other temperatures. Degradation rate of xanthopllylls, carotenes, and phycobilins was decreased in the materials treated at higher temperatures, showing a tendency of great pigment retention even for a long period storage. Chlorophyll il was retained more in sun dried materials whereas xanthophyll and carotene were retained more in the wet. 5. In order to maintain a higher pigment stability during storage, the materials must be treated at $60^{\circ}C$ in a dryer with rapid ventilation until the moisture level reaches about $6.0\%$. In case of treatment at a higher temperature than $60^{\circ}C$, they must be treated for a short period of time to minimize pigment degradation by heat. 6. The lavers treated with the solutions of pH around 7.8 showed a higher retention of chlorophyll a. When treated with the solutions of algal extracts such as agar and alginic acid, uniform layer of laver sheets and better surface gloss of dried layers were obtained.

• Journal of the Korean Electrochemical Society
• /
• v.6 no.1
• /
• pp.59-64
• /
• 2003

We fabricated mixed ionic-electronic conducting membranes, $CH_4\;Using\;{0.7}Sr_{0.3}Ga_{0.6}Fe_{0.4}O_{3-\delta}$, by solid state reaction method for solid oxide fuel cell. The membranes consisted of single perovskite phase and exhibited high relative density, $>95\%$. We coated $La_{0.6}Sr_{0.4}CoO_{3-\delta}$ layer using screen printing method in order to improve surface reactivity of the $La_{0.7}Sr_{0.3}Ga_{0.6}Fe_{0.4}O_{3-\delta}$. As a result, the oxygen permeation flux of the coated $La_{0.7}Sr_{0.3}Ga_{0.6}Fe_{0.4}O_{3-\delta}$ showed higher value, $0.5ml/min{\cdot}cm^2\;at\;950^{\circ}C$ than the uncoated one. Higher oxygen permeation was observed in the porously coated Lao $La_{0.7}Sr_{0.3}Ga_{0.6}Fe_{0.4}O_{3-\delta}$membranes with larger grain sizes. Syngas, $CO+H_2$, was successfully obtained from methane gas, $CH_4$, using the $La_{0.6}Sr_{0.4}CoO_{3-\delta}$ coated $La_{0.7}Sr_{0.3}Ga_{0.6}Fe_{0.4}O_{3-\delta}$, with over $40\%\;of\;CH_4$ conversion and syngas yield. $La_{0.7}Sr_{0.3}Ga_{0.6}Fe_{0.4}O_{3-\delta}$ membrane was stable even when it was exposed to the reducing environment, methane, for 600 hrs at $950^{\circ}C$.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.57 no.3
• /
• pp.238-247
• /
• 2012

To determine the possibility of solidified se wage sludge for use as a soil cover material in reclaimed land, the growth of energy crops and soil chemical properties investigated in each experimental plots during 2 years (2010 and 2011). The experimental plots consisted of the mixing with solidified sewage sludge plot (SS50), the covering with solidified sewage sludge plot (SS100), and the original reclaimed land plot (ORL) on reclaimed land for the intended landfill in Sudokwon Landfill Site Management Corporation (SLC). Plant height, measured in the second year (2011), was highest in the Geodae 1 grown at plots treated with solidified sewage sludge. The growth of energy crops cultivated in both SS50 and SS100 were better than in ORL. The contents of organic matter (OM) and total nitrogen (T-N) at both SS50 and SS100 were considerably higher than that of the ORL over 2 years. However, the soil from ORL showed higher salinity with high contents of exchangeable $Na^+$ cation than that of SS50 and SS100 over 2 years. We consider that soil chemical and physical properties on reclaimed land used in this study could be improved by the application of solidified sewage sludge due to following reasons. Firstly, the application of solidified sewage sludge may provide soil nutrients on reclaimed land i.e. the growth of energy crops better than in ORL, resulted in more OM and T-N contents in SS50 and SS100. Secondly, the top layers mixed or covered with solidified sewage sludge on reclaimed land may be prevented the salinity accumulation due to capillary rise to surface soil, and improved the cultivation layer for effectively propagating the rhizomes of energy crops. Thus the solidified sewage sludge may be a great soil cover materials for cultivation of bioenergy crops in reclaimed land.

• Economic and Environmental Geology
• /
• v.29 no.1
• /
• pp.65-75
• /
• 1996

Sulfate reduction and the precipitation of metal sulfides may have great potential to improve water quality of mine effluents in wetland treatment systems. Laboratory experiments using sulfate reducing bacteria (SRB) and limestone to treat effluents from the abandoned Dalsung tungsten-copper mine show that encouraging results, that have been attributed to sulfate reduction. Fe, Al, Cd, Cu and Zn are reduced to below detection limits with $99{\sim}100%$ metal removal rates, Mn is reduced by at least 90% to below 8.0 mg/l, and the pH is raised from 5.12 to 7.60 after 53 days of experiments. In the staged design, laboratory experiments are initiated to determine what would be reasonable substrate materials for remediation of the mine effluents. A substrate mixture containing 70% oak compost and 30% mushroom compost maintains $0.03{\sim}0.04mM$ of lactate, which provides good condition for the SRB granule. A downflow SRB wetland system is proposed as follows : 1) The lower part of the treatment system consists with a 25 cm thick layer of high quality (above 95% of $CaCO_3$) of limestone; 2) The geotextile (geonet) is recommended to be spread on the limestone bed to prevent clogging the limestones with the substrates; 3) The mixture of substrates with 70% oak and 30% spent mushroom composts, and SRB granules overlain on top of the geonet with 25 cm height. The sizes of the passive treatment systems are calculated according to metal loading and permeability criteria : 1) $220m^3$ ($15{\times}15{\times}1m$) for -1 level effluents; 2) $28m^3$ ($5.3{\times}5.3{\times}1m$) for -2 level; and 3) $2700m^3$ ($52{\times}52{\times}1m$) for the -3 level. The -3 level system needs to be broken down into 5 to 15 cells.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.35 no.2
• /
• pp.161-169
• /
• 1999

This paper describes the swimming and escaping behavior of amber fish, Seriola aureovittata released in the first bag net of the setnet and observed with telemetry techniques. The setnet used in experiment is composed of a leader, a fish court with a flying net and two bag nets having ramp net. The behavior of the fish attached an ultrasonic depth pinger of 50 KHz is observed using a prototype LBL fish tracking system. The 3-D underwater position ofthe fish is calculated by hyperbolic method with three channels of receiver and the depth of pinger. The results obtained are as follows: 1. The fish released on the sea surface was escaped down to 15 m depth and rised up to near the sea surface during 5 minutes after release. The average swimming speed of the fish during this time was 0.87 m/sec. 2. The swimming speed of the fish is decreased slowly in relation to the time elapsed and the fish showed some escaping behavior forward to the fish court staying 1 to 7 m depth layer near the ramp net. The average speed of the fish during this time was 0.52 m/sec. 3. During 25 minutes after beginning of hauling net, the fish showed a faster swimming speed than before hauling and an escaping behavior repeatedly from the first ramp net to the second one in horizontal. In vertical, the fish moved up and down between the sea surface and 20 m depth. After this time, the fish showed the escaping behavior forward to fish court after come back to the first ramp net in spite of the hauling was continued. It is found that the fish was escaped from the first ramp net to the fish court while the hauling was carried out. The average speed of the fish after beginning of hauling was 0.72 m/sec which increased 38.5 ％ than right before the hauling and showed 0.44 to 0.82 m/see of speed till escaping the first bag net. The average swimming speed during observation was 0.67 m/sec (2.2 times of body length).

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.39 no.1
• /
• pp.8-18
• /
• 2003

In order to investigate the oceanic condition of fishing ground in the adjacent sea of Naro Island, the oceanographic observation were carried out by the training vessel of Yosu National University on winter, spring, summer, and autumn in 2000. Main features in the observation are as follows; 1) the ranges of temperature, salinity, and chlorophyll-a were from 4.$3\circ_C$ to 10.$1\circ_C$, from 33.1 psu to 34.9 psu, and from 0.1 $ug$/$\Omega$ to 26.2 $ug$/$\Omega$ in winter, from 8.$1\circ_C$ to 13.$7\circ_C$, from 33.1 psu to 34.3 psu, and from 0.1 $ug$/$\Omega$ to 24.4 $ug$/$\Omega$ in spring, from 14.$5\circ_C$ to 24.$2\circ_C$, from 30.5 psu to 34.1 psu, and from 0.1 $ug$/$\Omega$ to 30.0 $ug$/$\Omega$ in summer, and from 14.$8\circ_C$ to 18.$6\circ_C$, from 30.1 psu to 34.0 psu, and from 0.1 $ug$/$\Omega$ to 19.1 $ug$/$\Omega$ in autumn, respectively, 2) the temperature in the coastal region was higher than that in the open ocean while salinity was lower, and the convection was identified between the surface and the bottom during in winter and autumn, and the thermocline were made between surface and 20m layer with vertical gradients of 4.$0\circ_C$/7m in summer, 3) the chlorophyll-a in the this region was varied in each season, being highly distributed in spring, on bottom and coastal region, and 4) an evidence of sea water intrusion toward Sori Island was observed, and of inner water intrusion from Yeoja Bay was observed.

• Korean Journal of Poultry Science
• /
• v.8 no.2
• /
• pp.55-68
• /
• 1981

Calcium and phosphorus are not only indispensable for the bone formation and body fluids equilibrium but also are major components of egg shell. It is nutritionally important, therefore, to investigate the metabolism of calcium and phosphorus and to search for optimum requirement of calcium and phosphorus and the availability of various sources of calcium an4 phosphorus by poultry. An attempt was made to review the nutrition of calcium and phosphorus in poultry diets. 1, Calcium and phosphorus have great interrelationship with vitamin D in their metabolisms. 2. Most of the plant-origin phosphorus are existing in phytic form and it leads to low availability when used in poultry rations, although calcium and phosphorus present in animal-origin or mineral supplements are highly available in general. 3. Calcium and phosphorus requirement from existing information indicated that 1.0% calcium and 0.7% phosphorus for broiler and egg-type chicks, and 3.5% calcium and 0.4% phosphorus for laying hen. 4. It has been recommended that calcium and phosphorus level should be increased when the feed intake was decreased or when the egg Production rate was higher or when the hens are old. 5. Mono-, ci-, tri-, calcium phosphate, calcium carbonate, bone meal, limestone and oyster shell u the most readily available among various sources of calcium phosphorus supplements. Soft rock phosphate, deflourinated phosphate and gypsum are somewhat inferior to the previous ones in bioavailability. 6. The effect of particle size of calcium supplements on egg shell quality and egg production rate is not yet clearly defined but recent works showed that oyster shell is more available when it was coarse and limestone is more available when it was fine in panicle. size. 7. Present data indicated that mixed feeding of oyster shell and limestone is superior to the single feeding of each on laying performance. 8. Significant interaction between phosphorus and sodium was observed, that is, excessive sodium decreased egg production in layer and body weight growth in broiler in the low phosphorus diets but increased them in the high phosphorus diets.

• Korean Journal of Soil Science and Fertilizer
• /
• v.13 no.2
• /
• pp.57-63
• /
• 1980

The effect of particle size of silicate fertilizer, crushed slag from the steel industry, on the behavior of silicate in soil was investigated through laboratory experiments. The silicate fertilizer was sieved to obtain three fractions of particles, coarser than 10 mesh 20-35 mesh, and finer than 100 mesh. Silicate concentration of the extract obtained by shaking 20 mg of particles, coarser than 10 mesh, 20-35 mesh, and finer than 100 mesh, in 50 ml of distilled water for 4 hours was 0.3, 1.0, and 3.2 ppm respectively. As shaking the mixture of the silicate fertilizer and soil proceeded, silicate concentration of the extract increased, and this increase after 4 hour shaking was attributed mainly to dissolution of soil silicate. When the mixture of soil and the silicate fertilizer was incubated under submerged condition, silicate concentration of the solution decreased for the first 2-4 weeks, thereafter increased with incubation time. During this incubation period, silicate concentration of the solution changed inversely with pH of the solution. After 6-10 weeks, however, both silicate concentration and pH of the solution increased with incubation time. Silicate concentration of the effluent from the 14.5 cm soil column of which top 4.5 cm was packed with the mixture of 30 g of soil and 30 mg of the silicate fertilizer reached maximum at 0.94 pore volumes for the particles of 20-35 mesh and 1.03 pore volumes for the particles finer than 100 mesh, whereas the effluent concentration reached maximum at 0.88 pore volumes for the soil column without the silicate fertilizer treatment. Soil analysis made after water percolation revealed that 1.5 pore volumes of water could leach down large amount of the water soluble silicate but not the sodium acetate extractable silicate, from top 3-6 cm soil layer.