• Journal of the Korean Institute of Landscape Architecture
• /
• v.40 no.4
• /
• pp.1-17
• /
• 2012

This study was carried out to propose the improved method by analyzing the change of green space arrangement and planting structure of apartment complexes in Seoul. 12 survey sites, which have obvious differences, were selected by reflecting the change of floor area ratio, underground parking place, and green space ratio. We divided the survey sites into four types that high green ratio(over 40%) apartment on natural ground, low green ratio(under 40%) apartment on natural ground, low green ratio(under 40%) apartment on artificial ground, and high green ratio(over 40%) apartment on artificial ground each period based on green space ratio and ground structure, plant crown volume, planting density, and planting pattern. The main factors of change of green space arrangement were green space ratio and ground structure. The Green space ratio was changed by the floor area ratio with constructing underground parking place and floor area ratio was adjusted by government policy and economic status. Average width of front green area has been changed from 10.0m in high green ratio apartment on natural ground for 3.5m, 2.7m, and 4.5m each period. The average width of the buffer green area has been changed from 15.0m in high green ratio apartment on natural ground of 7.7m, and 2.7m by extending parking place in the low green ratio apartment of artificial ground, so buffer green areas have been reduced and disconnected. So buffer green area in apartment complexes has been extended that the average width of the buffer green area was 3.8m caused by growing recognition of green since 2001. The ratio of native plant in canopy layer was increased from 45.1 % in the case of the high green ratio apartment of natural ground in 1980~1983 to 55.6%. Average plant crown volume increased from $1.27m^3/m^2$ in high green ratio apartment on natural ground for $3.47m^3/m^2$ in a low green ratio apartment on natural ground. But average plant crown volume is $0.27m^3/m^2$ in the high green ratio apartment of the artificial ground plant density of canopy layer was changed from 5 individuals per $100m^2$ to 14.5 individuals per $100m^2$. We should construct the buffer green area with natural ground and get the function of ecological and beautiful environment regarding to garden concept in case of front green area, width 4.5m. We should get the function of increasing green volume by multi-layer planting with shade woody species and flower woody species in case of back-side green area, width over 5.0m. We should get the function of covering the wall and increasing green landscape by planting with high woody species in case of side green area. We should apply the ecological planting technique to buffer green area and connect buffer green area to inner green area in apartment complexes.

• Korean Journal of Agricultural Science
• /
• v.10 no.2
• /
• pp.206-211
• /
• 1983

This study was carried out to compare the feed utilization between pelleted and all-mash diet of similar composition by growing chickens. Day-old broilers (Hubbard) and egg-type chickens(Hy-line) of commercial strain were employed in this experiment. The results obtained were summarized as follows. 1. The chickens fed pelleted diets were heavier than those of birds fed all-mash diets. The Hubbard broilers and Hy-line chickens fed pelleted diets weighed 2,702g and 812g respectively, at 9 weeks of age. In comparison, the Hubbard broilers and Hy-line chickens fed all-mash diets weighed 2,571g and 777g respectively, at 9 weeks of age. 2. The pellet-fed chickens consumed more feeds than birds fed all-mash diets in both types of strain. Feed efficiencies (gain/feed) of Hubbard and Hy-line chickens were 0.38 and 0.26 in pellet feeding groups, and 0.36 and 0.25 in all-mash feeding groups, respectively. The Hy-line chickens fed pelleted diets drank more water than birds fed all-mash diets. 3. Pellet feeding groups produced more dry matter excreta as compared with all-mash feeding groups, reflecting the pattern of feed consumption by these chickens. Nitrogen retention ratio of the Hubbard and Hy-line chickens were 57-67% and 65-73%, respectively. Chickens fed pelleted diets showed 1-4% higher nitrogen retention than chickens fed all-mash diets. 4. The ME/GE ratio of the Hubbard and the Hy-line at 8 weeks of age were 73.4-74.3% and 82.8-83.8%, respectively. Pellet feeding groups showed 1% higher ME/GE ratio than all-mash feeding groups. 5. The dietary productive energy calculated from respiratory quotient was $94.1-102.6kca/kg^{\frac{3}{4}}$ BW/day in pellet feeding groups. The ratios of PE/GE were 41.3-48.9% in pellet feeding groups and 39.0-45.8% in all-mash feeding groups. 6. It appears that pelleting the all-mash diet increases feed consumption and body weight gain of growing chickens. Feed efficiency and energy utilization were also improved by pelleting process. More research work should be done to establish the relationship clearly between feed pelleting and heat increments.

• Korean Journal of Poultry Science
• /
• v.18 no.2
• /
• pp.97-119
• /
• 1991

Effect of Streptococcus faecium(SF) and an antibiotic, Colistin(Col), supplemented to diets singly or in combination, on the performances and changes of intestinal population of microflora of broiler chicks studied. A total of 252, day-old chicks(Arbor Acre) of mixed sex(M：F＝1：1) were alloted into six groups. A diet with no Col and SF was referred as a control diet. The basal diets were added with two levels of SF, 0.04 and 0.08%, singly or in combination with Col 10ppm Another diet was prepared by adding only Col 10 ppm. Numbers of the microorganism in diets added with SF 0.04% and 0.08% were 7$\times$10$^{4}$ and 1.4$\times$10$^{5}$ /g diet respectively The diets consisting of corn and soybean meal as major ingredients were fed for a period of seven weeks . During the feeding trial, fresh excreta were sampled at the end of every week in a sterilized condition to count microbial changes from each dietary group. Microbial changes of large intestine were also measured from nine birds sacrificed at the end of the 4th and 7th weeks each time per dietary group. Excreta from all the groups were also collected quantitatively at the end of 3rd and 6th weeks to measure digestibility of the diets, At the end of 7th week, nine birds from each group were also sacrificed to measure weight changes of gastrointestinal tracts . Average body weight gains of broilers fed the diets added with SF 0.08% (2.37kg) or SF 0. 08%+col 10ppm(2.34kg) were significantly larger than that of the control(2.18kg). The weight gains of the other groups were not statistically different from that of the control Feed/gain ratios of the supplemental groups were better than that of control (P＜0.05) except that of birds fed the diet added only with SF 0.04%. Digestibilities of nutrients such as dry matter, crude protein, crude fat and total carbohydrates were not altered by the consumption of the diets added with SF and/or Col throughout the whole feeding period. As expected, the numbers of Streptococci in the excreta from birds fed diets added with SF increased significantly with a statistical difference between groups with SF 0.04% and SF 0.08% most of the time. However. addition of Colistin to the diets supplemented with SF did not give any effects on the number of the microorganism. Numbers of coliforms in the excreta were apparently reduced by feeding the diets added with SF and/or Col(P＜0.05). There were, however, no additive effects observed between the two feed additives in this regard when supplementing Col to the SF diets. Distributions of intestinal microflora exhibited exactly the same pattern as those of the excreta. Length of small intestine of the birds fed diets added with SF 0.08% with or without Col 10 ppm became significantly longer with a range of about 10% than those of the birds fed diets without SF. However, the empty weight of the small inestine of the former group was lighter than that of control These changes resulted in a significant reduction in weight/unit length of the intestine of the birds fed diets supplemented with Col and SF singly or in combination. In overall conclusion, diet added with SF 0.08% appeared most effective in improving broiler performances. Colistin added at a level of 10ppm was not beneficial at all in itself or in combination with SF in terms of broiler performances or changes of intestinal microflora population. The efficacy of SF and Col could be attributed to the changes of wall thickness of the small intestine.

• Journal of Wetlands Research
• /
• v.20 no.2
• /
• pp.155-160
• /
• 2018

We investigated to know the growth patterns of Aedes albopictus larva at the four different temperature conditions. Each of 120 individuals was placed into a $20m{\ell}$ vial and 12 sets (a set of 10 vials) were separated into 12 water tanks ($17{\times}24{\times}18cm^3$). Each water tank was composed of 3 the $1^{st}$ instar, 3 the $2^{nd}$ instar, 2 the $3^{rd}$ instar, and 2 the $4^{th}$ instar. Three sets of water tanks were placed under the four different incubator temperatures ($17^{\circ}C$, $21^{\circ}C$, $24^{\circ}C$, $28^{\circ}C$). We found that the eclosion rates were $20.00{\pm}5.77%$ at $21^{\circ}C$ and $3.33{\pm}3.33%$ at other temperatures. For the mosquito larva mortality rate, $1^{st}$ instar was $19.24{\pm}3.65%$, $2^{nd}$ instar was $16.48{\pm}3.25%$, $3^{rd}$ instar was $23.54{\pm}5.06%$, and $4^{th}$ instar was $40.74{\pm}7.08%$. The lowest mortality rate in growth stages according to temperature was $13.33{\pm}6.67%$ at $17^{\circ}C$ in $1^{st}$ instar larva, $7.41{\pm}7.41%$ at $21^{\circ}C$ at $2^{nd}$ instar larva, $10.74{\pm}6.43%$ at $24^{\circ}C$ in $3^{rd}$ instar larva, and $20.37{\pm}5.46%$ at $28^{\circ}C$ in $4^{th}$ instar larva. The survival period of mosquitoes in underwater were $26.33{\pm}0.67days$ at $17^{\circ}C$, $23.33{\pm}1.33days$ at $21^{\circ}C$, $20.00{\pm}2.52days$ at $24^{\circ}C$, and $11.67{\pm}1.20days$ at $28^{\circ}C$. From our results the most effective temperature to the normal growth of mosquito larva was $21^{\circ}C$, and the highest mortality rate was shown at the $4^{th}$ instar stage of larva growth. Our results would provide the basic data for the mosquito larva's growth pattern.

• The Journal of Dong Guk Oriental Medicine
• /
• v.4
• /
• pp.109-140
• /
• 1995

As we considered in the main subjects, investigations on the theory of 'Doctrine on five elements' motion and six kinds of natural factors(運氣學說)' through 'Wang Bing's Commentary(王氷 注本)' of 'The seven great chapters in The Yellow Emperor's Internal Classic Su Wen' ("黃帝內經素問 七篇大論") are as follows. (1) In The seven great chapters("七篇大論")' Wang Bing supplement theory and in the academic aspects as a interpreter, judging from 'forget(亡)' character. expressed in the 'The missing chapters("素問遺篇")', 'Bonbyung-ron("本病論")' and 'Jabeob-ron(刺法論)', 'The seven great chapters("七篇大論")' must be supplementary work by Wang Bing. Besides, he quoted such forty books as medical books, taoist books, confucianist books, miscellaneous books, etc in the commentary and the contents quoted in the 'Su Wen(素問)' and 'Ling Shu("靈樞")' scripture nearly occupy in the book. As a method of interpreting scripiure as scripture, he edited the order of 'Internal Classic("內經")' ascended from the ancient time and when he compensated for commentary, with exhaustive scholarly mind and by observing the natural phenomena practically and writing the pathology and the methods of treatment. We knew that the book is combined with the study of 'Doctrine on five elements motion and six kinds of natural factors(運氣學說)' (2) When we compare, analyze the similar phrase of 'The seven great chapters in The Yellow Emperor's Internal Classic Su Wen'("黃帝內經素問ㆍ七篇大論") through 'Wang Bing's Commentary(王氷 注本)', he tells abouts organized 'five elements(五行)' and 'heaven's regularly movement(天道運行)' rather than 'Emyangengsangdae-ron("陰陽應象大論")' in 'The seven great chapters("七篇大論")'. Also the 'Ohanunhangdae-ron("五運行大論")' because the repeated sentences with 'Emyangengsangdae-ron("陰陽應象大論")' is long they are omitted. And in the 'Youkmijidae-ron("六微旨大論")', 'Cheonjin ideology(天眞四象)' based on the 'Sanggocheonjin- ron("上古天眞論")', 'Sagijosindae-ron("四氣調神大論")' is written and in the 'Gigoupyondae-ron("氣交變大論")', the syndrome and symptom are explained in detail rather than 'Janggibeobsi-ron("藏氣法時論")', 'Okgijinjang-ron ("玉機眞藏論")' and in the 'Osangieongdae-ron("五常政大論")', the concept of 'five element(五行)' of the 'Gemgwejineon-ron("金櫃眞言論")' is expanded to 'the five elements' motion concept(五運槪念)' and in the 'Youkwonjeonggidae-ron("六元正紀大論")', explanations of 'The five elements' motion and six kinds of natural factors(運氣)' function are mentioned mainly and instead systematic pathology is not revealed rather than 'Emyangengsangdae-ron("陰陽應象大論")'. And in the 'Jijinyodae-ron("至眞要大論")', explanations of the change of atmosphere which correspond to treatment principle by 'The three Yin and Yang(三陰三陽)' as a progressed concepts are revealed. Therefore there are much similarity between the phrase of 'Emyangengsangdae-ron("陰陽應象大論")' and 'chapters of addition(補缺之篇)'. Generally, the doctrine which 'The seven great chapters("七篇大論")' are added by Wang Bing(王氷) is supported because there are more profound concepts rather than the other chapter in 'The seven great chapters("七篇大論")'. (3) When we study Wang Bing's(王氷) 'Pattern on five elements motion and six kinds of natural factors(運氣格局)' in 'The seven great chapter("七篇大論")', in the 'Cheonwongi-dae-ron("天元紀大論")', With 'Cheonjin ideology(天眞思想)' and the concepts of 'Owang(旺)'${\cdot}$'Sang(相)'${\cdot}$'Sa(死)'${\cdot}$'Su(囚)'${\cdot}$'Hu(休)' and 'Cheonbu(天符)'${\cdot}$'Sehwoi(歲會)' are measured time-spacially to the concept of 'Three Sum(三合)' the concept of 'Taeulcheonbu(太乙天符)' is explained. In the 'Ounhangdae-ron("五運行大論")', 'The calender Signs five Sum(天干五合)' is compared to the concepts of 'couples(夫婦)', 'weak-strong(柔强)' and in the 'Youkmijidae-ron("六微旨大論")', 'the relationship of obedience and disobedience(順逆關係)' which conform to the 'energy status(氣位)' change and 'monarch-minister(君相)' position is mentioned. In the 'Gikyobyeondae-ron("氣交變大論")', the concept of 'Sang-duk(相得)', 'Pyungsang(平常)' is emphasized but concrete measurement is mentioned. In the 'Osangieongdae-ron("五常政大論")', the detailed explanation with twenty three 'systemic of the five elements' motion(五運體系)' form and 'rountine-contrary treatment(正治. 反治)' with 'chill-fever-warm-cold(寒${\cdot}$熱${\cdot}$溫${\cdot}$凉)' are mentioned according to the 'analyse and differentiate pathological conditions in accordance with the eight principal syndromes(八綱辨證)'. In the 'Youkwonjeonggidae-ron("六元正紀大論")', Wang Bing of doesn't mention the concepts of 'Jungwun(中運)' that is seen in the original classic. In the new corrective edition, as the concepts of 'Jungwun, Dongcheonbu, Dongsehae and Taeulcheonbu(中運, 同天符, 同歲會, 太乙天符)' is appeared, Wang Bing seems to only use the concepts of 'Daewun, Juwun, and Gaekwun(大運, 主運, 客運)'. In the 'Jijinyodaeron("至眞要大論")', Wang Bing added detailed commentary to pathology and treatment doctrine by explaining the numerous appearances of 'Sebo, sufficiency, deficiency(歲步, 有餘, 不足)' and in the relation of 'victory-defeat(勝復)', he argued clearly that it is not mechanical estimation. (4) When we observe the Wang Bing's originality on the study of 'the theory of Doctrine on five elements' motion and six kinds of natural factors(運氣學說)', he emphasized 'The idea of Jeongindogi and Health preserving(全眞導氣${\cdot}$養生思想)' by adding 'Wang Bing's Commentary(王氷 注本)' of 'The seven great chapters("七篇大論")' and explained clearly 'The theory of Doctrine on five elements' motion and six kinds of natural factors(運氣學說)' and simpled and expanded the meaning of 'man, as a microcosm, is connected with the macrocosm(天人相應)' and with 'Atmosphere theory(大氣論)' also explained the meaning of 'rising and falling mechanism(升降氣機)'. In the sentence of 'By examining the pathology, take care of your health(審察病機 無失氣宜)'. he explained the meaning of pathology of 'heart-kidney-water-fire(心腎水火)' and suggested the doctrine and management of prescription. In the estimation and treatment, by suggesting 'asthenia and sthenia(虛實)' two method's estimation, 'contrary treatment(反治)' and treatment principals of 'falling heart fire tonifyng kidney water(降心火益腎水)', 'two class of chill and fever(寒熱二綱)' were demonstrated. There are 'inside and outside in the illness and so inner and outer in the treatment(病有中外 治有表囊)'. This sentence suggests concertedly. 'two class of superfies and interior(表囊二綱)' conforming to the position of disease. Therefore Wang Bing as an excellent theorist and introduced 'Cheoniin ideology(天眞思想)' as a clinician and realized the medical science. With these accomplishes mainly written in 'The theory of Doctrine on five elements' motion and six kinds of natural factors(運氣學說)' of 'The seven great chapters("七篇大論")', he interpreted the ancient medical scriptures and expanded the meaning of scriptures and conclusively contributed to the development of the study 'Korean Oriental Medicine(韓醫學)'.

• Journal of Acupuncture Research
• /
• v.17 no.2
• /
• pp.187-208
• /
• 2000

By the activation of ovary hormone, many morphological changes occur in the epithelial cell lines and muscle cells in rat uterus. These two cells in uterus are important to the implantation of embryo, maintaining pregnancy and starting parturition. One important change associated with the morphological change of these two cells in uterus is the change on prostaglandin(PG) metabolism. Its presence and synthesis in endometriurn and myometrium in uterus affects estrous cycle and the start of embryo implantation in uterus. It also performs as an important modulator in parturition. So the abnormally weak expression of PG causes difficulty during labor and over-expression causes pre-term labor. PG biosynthesis starts from either free or liberated arachidonic acids from membrane phospholipid by phospholipase. Such arachidonic acids are converted into PG catalyzed by Cyclooxygenase. Under normal physiological condition, Cyclooxygenase-1(COX-1) having 602 units of amino acids controls the synthesis of PG. It acts as a local hormone regulating vasomodulation of blood flow, flexible muscle movement, increasing the blood permeability and contributing the protective role in preserving integrity of the stomach lining and Cyclooxygenase-2 (COX-2) is induced by the inflammation, pregnancy and increased its expression until parturition. Lipid metabolite like PG is located in uterine and expression of COX-2 increased with pregnancy. Increased expression of COX proteins in epithelial cells and myometrial cells are told to increase the muscle contractility in uterus but decreased right after the labor in rat. It is a good sign indicating that COX proteins are deeply related to the start of labor. Currently, Several studies report the use of PG and COX-2 inhibitor as medication for controlled abortion or to prevent pre-term labor but they entail various side-effects. Our study proposed to suggest use of acupuncture as an another mediator to control abortion or pre-term labor without causing unnecessary side-effects by those medicines. Two acupuncture sites, LI-4 & SP-6 were selected due to their known efficacy. From the immunohistochemical staining of COX-2, normal expression of COX-2 protein in nonpregnant SD rat's uterus revealed that COX-2 protein was primarily detected in the lumina epithelial lining and in the epithelial cell lining contacting the stromal cells. High resolution optical microscopic scanning revealed distinguishable staining in the myometrial mucosa. LI-4 acupuncture administered nonpregnant rat's uterus showed strong expression for COX-2 in endometrium contacted with lumina epithelial lining of rat uterus and in myometrial mucosa. Stromal cells showed more staining than untreated nonpregnant rat's uterus and stronger staining in stromal cells contacting myometrial layer compared to untreated nonpregnant rat's uterus. SP-6 acupuncture administered nonpregnant rat's uterus showed weak expression for COX-2 in myometrial layers and stromal cells but no staining was visible in lumina epitheliai and glandular epithelial cells. Few stromal cells and myometrial mucosa were positively stained for COX-2. Pregnant SD rat's uterus was also immunostained for COX-2 expression after 18 days of pregnancy. Unlike to untreated nonpregnant rat's uterus, luminal epithelial cells were not positively stained for COX-2 but stronger staining for COX-2 was revealed in stromal cells. LI-4 acupunctured SD rat's uterus had very strong expression of COX-2 in luminal epithelial lining. Few stromal cells showed stronger positive COX-2 staining and myometrial layers also showed more expression than untreated pregnant rat. SP-6 acupuncture administered pregnant SD rat's uterus showed positive expression of COX-2 in epithelial cells of luminal mucosa layer but weaker than that of LI-4 acupuncture treatment's case. However, strong positive staining was revealed in stromal mucosa and myometrial layers. Virgin SD rat's uterus motility index during LI-4 acupuncture was 66.52 % (Prob〉T = 0.0197) compared to its motility before the acupuncture treatment but the motility index was slighdy elevated up to 79.58 % (Prob〉T = 0.1175) after the acupuncture. During the SP-6 acupuncture treatment for 30 minutes, uterus motility index was 90.52 % (Prob〉T = 0.1832) showing lesser decrement but consequently reached similar motility index decreasal to 79.95 % (Prob〉T = 0.0215) after the acupuncture treatment as LI-4 showed. LI-4 acupuncture tend to be a quick treatment to reducing the uterus motility in a virgin rat but eventually both two acupuncture administration created very similar reduction of uterus motility seeing the index after the both acupunctures. The uterus movement monitored during the LI-4 acupuncture administered for 30 minutes, Pregnant SD rat showed decreased motility down to 77.90 % (Prob〉 T = 0.0076) compared to uterus motility before the acupuncture and it continuously decreased down to 71.81 %(Prob〉T = 0.0214) after the removal of needle. The statistical analysis using paired t-test showed significance difference for both two motility indexs at =0.05. SP-6 acupuncture administered to pregnant SD rat also had similar pattern of decreasing uterus motility index down to 74.70 % (Prob〉T = 0.1730) during the initial 30 minutes acupuncture administration and it was continuously lowered to 71.52 % (Prob〉T = 0.0155) after the acupuncture. The paired t-test resuit for SP-6 suggest prompt response of uterus motility index to the SP-6 acupuncture treatment but consequently reached same level of inducing the motility reduction as LI-4 at =0.05 level.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.8 no.1
• /
• pp.1-9
• /
• 2006

Regardless of the recent observed warmer winters in Korea, more freeze injuries and associated economic losses are reported in fruit industry than ever before. Existing freeze-frost forecasting systems employ only daily minimum temperature for judging the potential damage on dormant flowering buds but cannot accommodate potential biological responses such as short-term acclimation of plants to severe weather episodes as well as annual variation in climate. We introduce 'dormancy depth', in addition to daily minimum temperature, as a complementary criterion for judging the potential damage of freezing temperatures on dormant flowering buds of grape vines. Dormancy depth can be estimated by a phonology model driven by daily maximum and minimum temperature and is expected to make a reasonable proxy for physiological tolerance of buds to low temperature. Dormancy depth at a selected site was estimated for a climatological normal year by this model, and we found a close similarity in time course change pattern between the estimated dormancy depth and the known cold tolerance of fruit trees. Inter-annual and spatial variation in dormancy depth were identified by this method, showing the feasibility of using dormancy depth as a proxy indicator for tolerance to low temperature during the winter season. The model was applied to 10 vineyards which were recently damaged by a cold spell, and a temperature-dormancy depth-freeze injury relationship was formulated into an exponential-saturation model which can be used for judging freeze risk under a given set of temperature and dormancy depth. Based on this model and the expected lowest temperature with a 10-year recurrence interval, a freeze risk probability map was produced for Hwaseong County, Korea. The results seemed to explain why the vineyards in the warmer part of Hwaseong County have been hit by more freeBe damage than those in the cooler part of the county. A dormancy depth-minimum temperature dual engine freeze warning system was designed for vineyards in major production counties in Korea by combining the site-specific dormancy depth and minimum temperature forecasts with the freeze risk model. In this system, daily accumulation of thermal time since last fall leads to the dormancy state (depth) for today. The regional minimum temperature forecast for tomorrow by the Korea Meteorological Administration is converted to the site specific forecast at a 30m resolution. These data are input to the freeze risk model and the percent damage probability is calculated for each grid cell and mapped for the entire county. Similar approaches may be used to develop freeze warning systems for other deciduous fruit trees.

• 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.

• Journal of Intelligence and Information Systems
• /
• v.19 no.1
• /
• pp.95-110
• /
• 2013

Recently, the amount of unstructured data being generated through a variety of social media has been increasing rapidly, resulting in the increasing need to collect, store, search for, analyze, and visualize this data. This kind of data cannot be handled appropriately by using the traditional methodologies usually used for analyzing structured data because of its vast volume and unstructured nature. In this situation, many attempts are being made to analyze unstructured data such as text files and log files through various commercial or noncommercial analytical tools. Among the various contemporary issues dealt with in the literature of unstructured text data analysis, the concepts and techniques of opinion mining have been attracting much attention from pioneer researchers and business practitioners. Opinion mining or sentiment analysis refers to a series of processes that analyze participants' opinions, sentiments, evaluations, attitudes, and emotions about selected products, services, organizations, social issues, and so on. In other words, many attempts based on various opinion mining techniques are being made to resolve complicated issues that could not have otherwise been solved by existing traditional approaches. One of the most representative attempts using the opinion mining technique may be the recent research that proposed an intelligent model for predicting the direction of the stock index. This model works mainly on the basis of opinions extracted from an overwhelming number of economic news repots. News content published on various media is obviously a traditional example of unstructured text data. Every day, a large volume of new content is created, digitalized, and subsequently distributed to us via online or offline channels. Many studies have revealed that we make better decisions on political, economic, and social issues by analyzing news and other related information. In this sense, we expect to predict the fluctuation of stock markets partly by analyzing the relationship between economic news reports and the pattern of stock prices. So far, in the literature on opinion mining, most studies including ours have utilized a sentiment dictionary to elicit sentiment polarity or sentiment value from a large number of documents. A sentiment dictionary consists of pairs of selected words and their sentiment values. Sentiment classifiers refer to the dictionary to formulate the sentiment polarity of words, sentences in a document, and the whole document. However, most traditional approaches have common limitations in that they do not consider the flexibility of sentiment polarity, that is, the sentiment polarity or sentiment value of a word is fixed and cannot be changed in a traditional sentiment dictionary. In the real world, however, the sentiment polarity of a word can vary depending on the time, situation, and purpose of the analysis. It can also be contradictory in nature. The flexibility of sentiment polarity motivated us to conduct this study. In this paper, we have stated that sentiment polarity should be assigned, not merely on the basis of the inherent meaning of a word but on the basis of its ad hoc meaning within a particular context. To implement our idea, we presented an intelligent investment decision-support model based on opinion mining that performs the scrapping and parsing of massive volumes of economic news on the web, tags sentiment words, classifies sentiment polarity of the news, and finally predicts the direction of the next day's stock index. In addition, we applied a domain-specific sentiment dictionary instead of a general purpose one to classify each piece of news as either positive or negative. For the purpose of performance evaluation, we performed intensive experiments and investigated the prediction accuracy of our model. For the experiments to predict the direction of the stock index, we gathered and analyzed 1,072 articles about stock markets published by "M" and "E" media between July 2011 and September 2011.

• Journal of Intelligence and Information Systems
• /
• v.24 no.4
• /
• pp.1-32
• /
• 2018

In addition to stakeholders including managers, employees, creditors, and investors of bankrupt companies, corporate defaults have a ripple effect on the local and national economy. Before the Asian financial crisis, the Korean government only analyzed SMEs and tried to improve the forecasting power of a default prediction model, rather than developing various corporate default models. As a result, even large corporations called 'chaebol enterprises' become bankrupt. Even after that, the analysis of past corporate defaults has been focused on specific variables, and when the government restructured immediately after the global financial crisis, they only focused on certain main variables such as 'debt ratio'. A multifaceted study of corporate default prediction models is essential to ensure diverse interests, to avoid situations like the 'Lehman Brothers Case' of the global financial crisis, to avoid total collapse in a single moment. The key variables used in corporate defaults vary over time. This is confirmed by Beaver (1967, 1968) and Altman's (1968) analysis that Deakins'(1972) study shows that the major factors affecting corporate failure have changed. In Grice's (2001) study, the importance of predictive variables was also found through Zmijewski's (1984) and Ohlson's (1980) models. However, the studies that have been carried out in the past use static models. Most of them do not consider the changes that occur in the course of time. Therefore, in order to construct consistent prediction models, it is necessary to compensate the time-dependent bias by means of a time series analysis algorithm reflecting dynamic change. Based on the global financial crisis, which has had a significant impact on Korea, this study is conducted using 10 years of annual corporate data from 2000 to 2009. Data are divided into training data, validation data, and test data respectively, and are divided into 7, 2, and 1 years respectively. In order to construct a consistent bankruptcy model in the flow of time change, we first train a time series deep learning algorithm model using the data before the financial crisis (2000~2006). The parameter tuning of the existing model and the deep learning time series algorithm is conducted with validation data including the financial crisis period (2007~2008). As a result, we construct a model that shows similar pattern to the results of the learning data and shows excellent prediction power. After that, each bankruptcy prediction model is restructured by integrating the learning data and validation data again (2000 ~ 2008), applying the optimal parameters as in the previous validation. Finally, each corporate default prediction model is evaluated and compared using test data (2009) based on the trained models over nine years. Then, the usefulness of the corporate default prediction model based on the deep learning time series algorithm is proved. In addition, by adding the Lasso regression analysis to the existing methods (multiple discriminant analysis, logit model) which select the variables, it is proved that the deep learning time series algorithm model based on the three bundles of variables is useful for robust corporate default prediction. The definition of bankruptcy used is the same as that of Lee (2015). Independent variables include financial information such as financial ratios used in previous studies. Multivariate discriminant analysis, logit model, and Lasso regression model are used to select the optimal variable group. The influence of the Multivariate discriminant analysis model proposed by Altman (1968), the Logit model proposed by Ohlson (1980), the non-time series machine learning algorithms, and the deep learning time series algorithms are compared. In the case of corporate data, there are limitations of 'nonlinear variables', 'multi-collinearity' of variables, and 'lack of data'. While the logit model is nonlinear, the Lasso regression model solves the multi-collinearity problem, and the deep learning time series algorithm using the variable data generation method complements the lack of data. Big Data Technology, a leading technology in the future, is moving from simple human analysis, to automated AI analysis, and finally towards future intertwined AI applications. Although the study of the corporate default prediction model using the time series algorithm is still in its early stages, deep learning algorithm is much faster than regression analysis at corporate default prediction modeling. Also, it is more effective on prediction power. Through the Fourth Industrial Revolution, the current government and other overseas governments are working hard to integrate the system in everyday life of their nation and society. Yet the field of deep learning time series research for the financial industry is still insufficient. This is an initial study on deep learning time series algorithm analysis of corporate defaults. Therefore it is hoped that it will be used as a comparative analysis data for non-specialists who start a study combining financial data and deep learning time series algorithm.