• Journal of Korean Society for Quality Management
• /
• v.5 no.2
• /
• pp.59-120
• /
• 1977

An antibiotic 'P', which is one of the products of the Gist Brocades N. V. is being tested by its research department as fungicide on seed-potatoes. For this testing they designed experiments, with two control groups, one competitor's product, eight formulations of the antibiotic to be tested in different concentrations and one mercury treatment which can not be used in practice. The treated potatoes were planted in three different regions, where bifferent conditions prevail. After several months the harvested potatoes are divided in groups according to their diameter, potato illness is analysed and counted. These data were summarised in percentage and given to us for Analysis. We approached and analysed the data by following methods: a. Computation of the mean and standard deviation of the percenage of good results in each size group and treatment. b. Computation of the experimental errors by substraction of each treatment mean from observed data. c. Description of the frequency table, plotting of a histogram and a normal curve on same graph to check normality. d. Test of normality paper and chi-sqeare test to check the goodness of fit to a normal curve. e. Test for homogeneity of variance in each treatment with the Cochran's test and Hartley's test. f. Analysis of Variance for testing the means by one way classifications. g. Drawing of graphs with upper and lower confidence limits to show the effect of different treatments. h. T-test and F-test to two Control mean and variance for making one control of Dunnett's test. i. Dunnett's Test and calculations for numerical comarision of different treatments wth one control. In region R, where the potatoes were planted, it was this year very dry and rather bad conditions to grow potatoes prevailed during the experimental period. The results of this investigation show us that treatment No.2, 3 and 4 are significantly different from other treatments and control groups (none treated, just like natural state). Treatment no.2 is the useless mercury formulation. So only No. 3 and 4, which have high concentrations of antibiotic 'P', gave a good effect to the potatoes. As well as the competitors product, middle and low concentrated formulations are not significantly different from control gro-ups of every size. In region w, where the potatoes got the same treatments as in region R, prevailed better weather conditions and was enough water obtainable from the lake. The results in this region showed that treatment No. 2, 3, 4, and 5 are Significantly different from other treatments and the control groups. Again No.2 is the mercury treatmentin this investigation. Not only high concentrated formulation of antibiotic 'P', but also the competitor's poroduct gave good results. But, the effect of 'P', was better than the competitors porduct. In region G, where the potatoes got the same treatments as in the regions R and w. and the climate conditions were equal to region R, the results showed that most of the treatments are not significantly different from the control groups. Only treatment no. 3 was a little bit different from the others. but not Significantly different. It seems to us that the difference between the results in the three regions was caused by certain conditions like, the nature of the soil the degres of moisture and hours of sunshine, but we are not sure of that. As a conclusion, we can say that antibiotic 'P' has a good effect on potatoes, but in most investigations a rather high concentration of 'P' was required in formulations.

• Journal of radiological science and technology
• /
• v.8 no.2
• /
• pp.21-28
• /
• 1985

We serveyed the actual condition of business of the departments of radiology of 45 health conters (except 3) in the area of Seoul, Kyungki and Inchon from March, 1984 to November, 1984. The results are summarized as follows : 1. T.O. of the radiologic technologist is three persons in each health center of Seoul area, and one person in each one of Kyungki and Inchon area. P.O. is 2-5 persons in Seoul area, 1-2 persons in Kyungki or Inchon area. 2. The number of all the radiologic technologists employed now is 75 persons, and among all of them, when analized by position class 7th is 54.7%, class 8th 28.0%, class 9th is 13.3%, and class 6th is 2.7%, and by sex, female is 68.0%, male is 32.0%, by educational background, for the most part, junior college graduates come to 73.3%, by age group 60% of them is in their twenties, 16.0% is in their thirties and forties, 8.0% is in their fifties, and by career after certificate 60% have the career of 1-5 years, 13.3% have the one of 6-7 years or mor than 21 years, and 6.7% have the one of 11-15 years of 16-20 years. 3. All the diagnostic x-ray equipment being kept is 62, and among them flxing equipment is 71.0%, portable equipment is 29.0% and by rating of X-ray equipment, maximum tube current 100 mA is 46.8%, maximum KV 100KVP is 72.6%, the most part. 4. Photofluorographic camera and hood are equipped in every health center. While, as to the radiographic cassettes, $14{\times}14"$ cassetts are equipped in every health center, but cassettes of other sizes are in half of them. 5. Bucky's table is equipped in 11.9% health centers, the automatic processor is in 21.4%, the photofluorographic film changer is 9.5%, the grid is 73.8%, the protective apron is in 88.1%, and the protective glove is in 57.1% health centers. 6. The number of the people who got the x-ray examination for one year (by the year 1989) is the most, 1,000-6,000 in direct radiography of the chest, or 15,0001-45,000 in the health centers of Seoul area, 5,000-20,000 in Kyungki and Inchon area in photofluorography of the chest. Moreover, other radiographies are being taken extremely limitedly in all health centers. 7. In processing types of x-ray film, automatic processing is used in 9 health centers (21.4%), manual tank processing is in 30 (71.4%), and manual tray processing in 3 (7.2%). 8. As for collimation of x-ray exposure field, "continual using restricted by a subject size" has the most part, 78.6% "restricted using at every radiography" has 19%, and the case of "never considered" has 2.4% response. 9. As for the dosimeter used for radiation control, film badge (35.7%) and pocket dosimeter (26.2%) are used, and in 38.1% health centers the dosimeter is not equipped at all. Consideration of the previous radiation exposure is being done in only one health center. 10. Reading of radiographs is mainly depended on the radiologists electively (45.2%) or on the genral practitioners(45.2%).

• The Journal of Korean Academy of Prosthodontics
• /
• v.46 no.5
• /
• pp.479-489
• /
• 2008

Statement of the problem: Under anatomical limitations on maxillary posterior region, a poor crown-to root ratio acting on dental implants can result in undesirable stress in surrounding bone, which in turn can cause bone defects and eventual failure of implants. Purpose: The purpose is to compare stress distribution due to different crown-root ratio and effect of splinting between natural teeth and implants in maxillary molar area under different loads. Material and methods: Analysis of stress arising supporting bone of the natural teeth and the implant was made with 3-dimensional finite element method. The model simulated naturel teeth was made with 2nd premolar and 1st molar in the maxillary molar region (Model T). The model simulated implants placed on same positions with two parallel implants of Straumann Dental Implant cemented abutment (Model I). Each model was designed in different crown-root ratio (0.7:1, 1:1, 1.25:1) and set cement type gold crown to make it non-splinted or splinted. After that, 300 N force was loaded to each model in five ways (Load 1: middle of occlusal table, Load 2: middle of buccal cusp, Load 3: middle of lingual cusp, Load 4: horizontal load to buccal cusp of anterior abutment only, Load 5: horizontal load to middle of buccal cusp of each abutment), and stress distribution was analyzed. Results and conclusion: On all occasions, stress was concentrated at the cervical region of the implant. Under load 1, 2 and 3, stress was not increased even when crown-root ratio increases, but under load 4 and 5, when crown-root ratio increases, stress also increased. There was difference in stress values between natural teeth and implants when crown-root ratio gradually increases; In case of natural teeth, splinting decreased stress under vertical and horizontal loads. In case of implants, splinting decreased stress under vertical loads 1,2 and 3, but increased maximal stress under loads 2 and 3. Under horizontal loads, splinting decreased stress, however the effect of splinting decreased under load 5 than load 4. Furthermore, the stress was increased, when crown-root ratio is 1.25:1. Clinical implications: This limited finite element study suggests that the stress on supporting bone may be increased under non-axial loads and poor crown-root ratio. Under poor crown-root ratio, excessive stress was generated at the cervical region of the implant, and decreased splinting effect for stress distribution, which can be related to clinical failure.

• Journal of Korean Society of Forest Science
• /
• v.57 no.1
• /
• pp.8-13
• /
• 1982

The present study was conducted during the period July 18-August 11, 1981, at the four industrial complexes(Pohang, Ulsan, Onsan, and Yeochon) using the naked eye method. The study included the survey of plant species growing within a 2km from the area of pollution source, selection of plants tolerant to air pollution and effects of pollution on their growth. The results obtained are as follows; 1) Vascular plants surveyed as to industrial complexes consist of 90 families, 6 subfamilies, 213 genera and 290 species. Major plant families and species for each area are as follows: 1) Pohang area: 54 families, 102 genera and 126 species Leguminosae(16) Graminae(11), Compositae(9) Rosaceae(9) 2) Ulsan area: 61 families, 131 genera, and 158 species Compositae(17) Gramineae(16) Leguminosae(15) Rosaceae(9) Polygonaceae(9) Solanaceae(6) 3) Onsan area: 46 families, 98 genera, and 158 species Compositae(12) Gramineae(12) Leguminosae(12) Rosaceae(6) Polygonaceae(6) 4) Yeochon: 71 families, 150 genera, and 188 species Rosaceae(15) Leguminosae(15) Compositae(14) Graminese(14) Polygonaceae(8) Salicaceae(6) 2. Table 1 shows tolerant plants to air pollution in this study. 3. The following plants may be recommended as planting species at the industrial complex area. Leguminosae: Amorpha fruticosa L. Lespedeza bicolor Turcz. Lespedeza maximowiczii Schneider Robinia pseudoacacia L. Sophora japonica L. Fagaceae: Quercus acutissima Carr. Quercus serrata Thunb. Quercus variabilis Bl. Salicaceae: Populus tomentiglandulosa T. Lee Salix Pseudo-lasiogyne Leveille Oleaceae: Ligustrum japonicum Thunb. Ligustrum obtusifolumn S. et Z. Betulaceae: Alnus hirsuta (Space) Ruprecht Celastraceae: Euonymus japonicus thunb. Cparifolicceae: Viburnum awabucki K. Koch. Pinaceae: Cedrus deodara Loundon.

• Journal of Global Scholars of Marketing Science
• /
• v.18 no.4
• /
• pp.33-59
• /
• 2008

This paper aims to analyse the preference for German farm tourism among the German population. For this reason, we conducted an empirical study in Germany during summer 2007 and we applieda structural equation model based on partial leasts quares(PLS) to analyse the data. In the following chapters we will introduce the literature review and our conceptual frame work. We will then outline the procedures we adopted and the results of the empirical analysis. In the final part so me conclusions will be presented and a discussion will follow in order to draw the future directions of our research. According to our hypotheses, the possibility that agri-tourism enters in the evoked set of an individual is higher: H1: The higher the information degree about it. H2: The lower the influence of the social stimuli. H3: The higher the physical exposure to it (experience). H4: The higher the wellness image of agri-tourism. H5: The higher the traditional image of agri-tourism. H6: The higher the exciting image of agri-tourism. H7: The higher the perceived value for money. Our further hypotheses affirm that the possibility that agri-tourism enters in the evoked set of an individual is higher: H8: The lower the perceived risk. H9: The higher the motive to enjoy a holiday in the nature. H10: The higher the motive to enjoy a sport holiday. H11: The lower the motive to have an organized holiday. H12: The lower the motive to have a holiday abroad. H13: The lower the motive of action and night life. H14: The higher the motive to spend a holiday with the family. H15: The lower the motive to spend a city holiday. Finally, our model has some socio-demographics data. As we mentioned before, German agri-tourism has traditionally been the travel destination of large-size families, with low-to-middle income. For that reason, our final hypothesises are the following: the possibility that agri-tourism enters in the evoked-set of an individual is higher: H16: The higher the number of family members. H17: The lower the family income. Since in this study we use a path model with a PLS approach, we are able to state some interrelations among the exogenous latent variables: H18: The motive of sport holiday has a positive influence towards nature motives. H19: The physical exposition to agri-tourism has a positive influence toward information. H20: The motive of family holiday has a negative influence toward the motive of action and night life. H21: Social stimuli have a positive influence towards individuals risk perceptions. H22: Social stimuli have negative influence towards experience. Data for this study were gathered via administrated questionnaires during the summer 2007 within the frame of an academic "marketing research" course. The corresponding t-values are assessed using the bootstrapping method with 500 re-samples. In our model 61% of the degree of appreciation of German agri-tourism (evoked set) is explained by five independent variables: value for money ($0.335^{{\ast}{\ast}{\ast}}$) (H7) experience ($0.267^{{\ast}{\ast}}$) (H3), exciting image ($0.204^{\ast}$) (H6) organisation ($-0.162^{\ast}$) (H11) and holiday abroad ($-0.156^{\ast}$) (H12). The variance explained ($R^2$) for the other endogenous variables are the following: nature 24.3%, information 14.1%, action holiday 13.8%, risk perception 5.8% and experience 2.4%. An overview can be inferred from table 5. The results also allow us to test each of the proposed hypotheses. With exception of organization and abroad, none of the others travel style factors (H9 to H15) seem to have any significant impact towards evoked set which leads to the rejection of the respective hypotheses. As expected, social stimuli have a significant influence on individuals' risk perception (H21 accepted), however neither the former nor the latter have a valuable impact on evoked set (rejection of H2 and H8). Besides, since the influence of social stimuli towards experience is not significant, also H22 has to be rejected. Experience influences information (H19 accepted) but the latter does not affect significantly the evoked set (H1 rejected). Both H4 as well as H5, referring respectively to the perceived images of German agri-tourism as a wellness destination and the traditional image of the German farm tourism have to be rejected. Finally, none of the demographic data included in the model explains significantly the variance of the factor evoked set. Therefore neither H16 nor H17 has been accepted. As far as the interrelation between sport and nature (H18) and family and action (H20) are concerned, the stated relationship among these variables has been statistically confirmed. Our path model based on partial least squares shows the factors influencing the preference for farm tourism in Germany. Among others value for money and experience are the most significant ones. Practical implications are discussed.

• Journal of Nutrition and Health
• /
• v.1 no.2
• /
• pp.61-85
• /
• 1968

The experiment was carried out on eight healthy male subjects of 19-22 years old and they were maintained The experiment was carried out on eight healthy male subjects of 19-22 years old and they were maintained under the rice diet, the general mixed diet and the high animal protein diet for 7 days of each experimental period, respectively. The composition of each diet was indicated in Table 3. The allowance' of each nutrient per head per day in each diet were as follows. a) In the rice diet; protein:139.5 g (animal source:64.6 g, vegetable source:74.9 g), fat:25.3 g (animal source:11.3 g, vegetable source:14.0 g), carbohydrate: 644.0 g, calcium:982.3 mg, phosphorus: 2369.2mg, thiamine:2.1 mg, riboflavin: 1.6 mg, and 3211 Cal. b) In the general mixed diet; protein: 97.4g (animal source: 10.5 g, vegetable source: 86.9 g), fat:40.3 g (animal source:7.1 g, vegetable source:33.2 g), carbohydrate:620.7 g, calcium:887.1m g, phosphorus:2200.8m g, thiamine: 1.8m g, riboflavin:1.9m g, and 3158 Cal. c) In the high animal protein diet; protein: 135.6 g (animal source:68.8 g, vegetable source:66.8 g), fat:32.5g (animal source:17.9 g, vegetable source:14.6 g), carbohydrate:930.9 g, calcium: 626.0 mg, phosphorus: 1998.9 mg, thiamine: 1.5 mg, riboflavin: 1.5 mg, and 3194 Cal. The absorption rates of protein, fat, carbohydrate, calcium, phospherus, thiamine and riboflavin in each diet were observed. The results obtained were summarized as follows: 1. The absorption rates of protein and fat in the rice diet were 88.t % and 71.3.%. 2. The absorption rates of protein and fat in the general mixed diet were 83.4% and 86.4%. 3. The absorption rates of protein and fat in the high animal protein diet were 86.8% and 82.4%. 4. The nitrogen balances in the rice, the general mixed and the high animal protein diet groups were +5.7g, +2.3g and +4.0g respectively. 5. The absorption rates of carbohydrate in each diet were all above 95 % and so seemed to be almost completely absorbed except the cellulose in the diets. 6. The calory utilization rates in each diet were all above 93 %. 7. The minimum absorption rates of protein, fat and calory were 83.8%, 63.3%, and 89.7% in the rice diet, 80.0%, 80.9% and 85.9%, in the general mixed diet and 83.6%, 75.7% and. 89.3 % in the high animal protein diet respectively. Therefore, it is assumed that these data might be significant in practical use. 8. The protein absorption rates of the rice diet (88.1 %) was better than that of the general mixed diet (83.4 %). 9. The fat absorption rates of the general mixed diet and the high animal protein diet (86.4% and 82.4%) were signficantly better than that of the rice diet (71.3%). 10. The calcium absorption rates of each diet, rice diet, the general mixed diet and the high animal protein diet were 41.6%, 36.0%, and 27.7%, respectively. 11. The phosphorus absorption rates of each diet were 51.4%, 56.0% and 52.3%, respectively.12. The phosphorus absorption rate seemed better than that of calcium. 13. The thiamine absorption rates of each diet seemed 27.0 %, 42.9 % and 29.5 %, respectively. 14. The riboflavin absorption rates of each diet seemed above 30.6%, 27.1%, and 39.3% respectively. 15. The excretion amounts of thiamine or riboflavin were much more than the amounts ingested of the both vitamins. Therefore, the certain amount of both vitamins seemed to synthesize in the intestine.

• Journal of Animal Science and Technology
• /
• v.45 no.5
• /
• pp.703-710
• /
• 2003

The purpose of this study was to detect association between genetic variation and economic trait in the porcine heart type fatty acid-binding protein gene as a candidate gene for the traits related with growth and meat quality in pigs. The H-FABP is a 15-kDa protein expressed in several tissues with high demand for fat metabolism such as cardiac and skeletal muscle and lactating mammary gland. H-FABP is small intracellular protein involved in fatty acid transport from the plasma membrane to the site of $\beta$-oxidation and/or triacylglycerol or phospholipid synthesis. In this study, H-FABP PCR-RFLP was performed in F$_2$ population composed of 214 individuals from an intercross between Korean Native Boars and Landrace sows. PCR products from two primer sets within H-FABP gene were amplified in 850bp and 700bp. Digestion of PCR products with the restriction digestion enzymes HaeⅢ and HinfⅠ, revealed fragment length polymorphisms(RFLPs). The genotype frequencies from H-FABP/HaeⅢ was .29 for genotype DD, .53 for genotype Dd, and .15 for genotype dd, respectively. The genotype frequencies of HH, Hh, and hh from H-FABP/HinfⅠ was .38, .41 and .20, respectively, in the population. Relationships between their genotypes and economic traits were estimated. In H-FABP/HaeⅢ locus, there were specific genotypes(Dd and dd) associated with economic traits such as body weights at 3, 5, 12, and 30 week of age (p〈.05 to .001). The ‘d’ allele was associated with gaining of body weight. In H-FABP/HinfⅠ locus, Genotypes of HH and Hh associated with growth traits such as body weights at 5, 12, and 30 week of age (p〈.05 or p〈.001) and back fat thickness, body fat including abdominal and trimmed fat (p〈.001) and intramuscular fat(p〈.05) The ‘H’ allele was positively associated with gaining of body weight and fatness deposition. In conclusion, a significant association of the H-FABP gene from its genetic variation was found on body weight, intramuscular fat and backfat thickness.

• Journal of The Korean Society of Inherited Metabolic disease
• /
• v.12 no.2
• /
• pp.85-93
• /
• 2012

Purpose: Developmental delay is caused by very diverse etiologic diseases. Most chronic disorders has some influence on development. Chronic or acute disorders of CNS are main etiologic diseases of developmental delay. Up to now, over 60 diseases are included in organic acidopathies and most of them causes acute or chronic recurrent CNS damage and developmental delay. We have done this study to find out the importance of organic acidopathies causing developmental delay in Korean childhood and adolescent patients. Method: Retrograde analysis for 738 patients with developmental delay whose clinical informations are available and have done urine organic acid analysis for 5 years period, between Jan. 1st 2007 to Dec. 31th 2011. Statistical analysis was done with Student's t test using SPSS. Result: Out of 738 patients, 340 patients (46.1%) showed abnormalities on urine organic acid analysis. The most frequent disease was mitochondrial respiratory chain disorders (MRCD) (253, 34.3%), followed by ketolytic defects(39, 5.3%), 3-hydroxyisobutyric aciduria (26, 3.5%), glutaric aciduria type II (8, 1.1%), pyruvate dehydrogenase deficiency (3, 0.4%), 3-methylglutaric aciduria (2, 0.3%), glutaric aciduria type I (2, 0.3%), ethylmalonic aciduria (1, 0.15%), methylmalonic aciduria (1, 0.15%), HMG-CoA lyase deficiency (1, 0.15%), 3-methylcrotonylglycinuria (1, 0.15%), fatty acid oxidation disorders(1, 0.15%) and FAOD (1, 0.15%). Conclusion: Mitochondrial disorders are most frequent etiologic disease on all age group, followed by ketolytic defects and various organic acidopathies. The number and diversities of organic acidopathies emphasize meticulous evaluation of basic routine laboratory examinations and organic acid analysis with initial sample on every developmental patient.

• Korean journal of applied entomology
• /
• v.37 no.2
• /
• pp.143-149
• /
• 1998

Development and life table statistics of indian meal moth (Plodia interpunctella, Hiibner)on brown rice (Sativa oryzae L., var 'Ilpum' ) were tested under five different temperatures (17, 20, 25,28 and 3 2 f 0.5"C). The developmental response of females to the temperature was not significantlydifferent from that of males. In the tested temperature range, developmental period and life span of adultmoth decreased as the temperature increased and ranged from 149.9f30.4 to 38.1 k5.6 days and from19.4f 5.1 to 6.9k2.0 days at 17$^{\circ}$C and 32"C, respectively. Emergence rate increased with the increaseof temperature and ranged from 13.0f 6.2% at 17$^{\circ}$C to 49.2f 25.9% at 32$^{\circ}$C. However, hatching ratecurve in relation to the temperature was dome shape with the peak of 73.8 k5.37~a t 25"C, suggesting thathatching is inhibited by high temperature above that temperature. As the temperature increased, femalesconcentrated their oviposition on the second day after emergence. In the temperature range of 17 SIM 25"C, the number of eggs laid per female were not related to the temperature and ranged from 133.4f 37.6to 154.3k57.4. But the number of eggs laid per female decreased at 32$^{\circ}$C which suggests closerelationship with hatching ability. The net reproduction rate was highest at 28$^{\circ}$C and followed by those at25$^{\circ}$C and 20$^{\circ}$C. However intrinsic rate of natural increase of the moth population on brown rice wasestimated to be highest at 32$^{\circ}$C (0.065 per day), probably due to the short developmental period, highemergence rate and the concentrated oviposition of females on earlier days of the emergence.ition of females on earlier days of the emergence.

• Journal of Intelligence and Information Systems
• /
• v.23 no.3
• /
• pp.139-153
• /
• 2017

Corporate bankruptcy can cause great losses not only to stakeholders but also to many related sectors in society. Through the economic crises, bankruptcy have increased and bankruptcy prediction models have become more and more important. Therefore, corporate bankruptcy has been regarded as one of the major topics of research in business management. Also, many studies in the industry are in progress and important. Previous studies attempted to utilize various methodologies to improve the bankruptcy prediction accuracy and to resolve the overfitting problem, such as Multivariate Discriminant Analysis (MDA), Generalized Linear Model (GLM). These methods are based on statistics. Recently, researchers have used machine learning methodologies such as Support Vector Machine (SVM), Artificial Neural Network (ANN). Furthermore, fuzzy theory and genetic algorithms were used. Because of this change, many of bankruptcy models are developed. Also, performance has been improved. In general, the company's financial and accounting information will change over time. Likewise, the market situation also changes, so there are many difficulties in predicting bankruptcy only with information at a certain point in time. However, even though traditional research has problems that don't take into account the time effect, dynamic model has not been studied much. When we ignore the time effect, we get the biased results. So the static model may not be suitable for predicting bankruptcy. Thus, using the dynamic model, there is a possibility that bankruptcy prediction model is improved. In this paper, we propose RNN (Recurrent Neural Network) which is one of the deep learning methodologies. The RNN learns time series data and the performance is known to be good. Prior to experiment, we selected non-financial firms listed on the KOSPI, KOSDAQ and KONEX markets from 2010 to 2016 for the estimation of the bankruptcy prediction model and the comparison of forecasting performance. In order to prevent a mistake of predicting bankruptcy by using the financial information already reflected in the deterioration of the financial condition of the company, the financial information was collected with a lag of two years, and the default period was defined from January to December of the year. Then we defined the bankruptcy. The bankruptcy we defined is the abolition of the listing due to sluggish earnings. We confirmed abolition of the list at KIND that is corporate stock information website. Then we selected variables at previous papers. The first set of variables are Z-score variables. These variables have become traditional variables in predicting bankruptcy. The second set of variables are dynamic variable set. Finally we selected 240 normal companies and 226 bankrupt companies at the first variable set. Likewise, we selected 229 normal companies and 226 bankrupt companies at the second variable set. We created a model that reflects dynamic changes in time-series financial data and by comparing the suggested model with the analysis of existing bankruptcy predictive models, we found that the suggested model could help to improve the accuracy of bankruptcy predictions. We used financial data in KIS Value (Financial database) and selected Multivariate Discriminant Analysis (MDA), Generalized Linear Model called logistic regression (GLM), Support Vector Machine (SVM), Artificial Neural Network (ANN) model as benchmark. The result of the experiment proved that RNN's performance was better than comparative model. The accuracy of RNN was high in both sets of variables and the Area Under the Curve (AUC) value was also high. Also when we saw the hit-ratio table, the ratio of RNNs that predicted a poor company to be bankrupt was higher than that of other comparative models. However the limitation of this paper is that an overfitting problem occurs during RNN learning. But we expect to be able to solve the overfitting problem by selecting more learning data and appropriate variables. From these result, it is expected that this research will contribute to the development of a bankruptcy prediction by proposing a new dynamic model.