• Journal of Intelligence and Information Systems
• /
• v.25 no.2
• /
• pp.123-139
• /
• 2019

The Korean NPL market was formed by the government and foreign capital shortly after the 1997 IMF crisis. However, this market is short-lived, as the bad debt has started to increase after the global financial crisis in 2009 due to the real economic recession. NPL has become a major investment in the market in recent years when the domestic capital market's investment capital began to enter the NPL market in earnest. Although the domestic NPL market has received considerable attention due to the overheating of the NPL market in recent years, research on the NPL market has been abrupt since the history of capital market investment in the domestic NPL market is short. In addition, decision-making through more scientific and systematic analysis is required due to the decline in profitability and the price fluctuation due to the fluctuation of the real estate business. In this study, we propose a prediction model that can determine the achievement of the benchmark yield by using the NPL market related data in accordance with the market demand. In order to build the model, we used Korean NPL data from December 2013 to December 2017 for about 4 years. The total number of things data was 2291. As independent variables, only the variables related to the dependent variable were selected for the 11 variables that indicate the characteristics of the real estate. In order to select the variables, one to one t-test and logistic regression stepwise and decision tree were performed. Seven independent variables (purchase year, SPC (Special Purpose Company), municipality, appraisal value, purchase cost, OPB (Outstanding Principle Balance), HP (Holding Period)). The dependent variable is a bivariate variable that indicates whether the benchmark rate is reached. This is because the accuracy of the model predicting the binomial variables is higher than the model predicting the continuous variables, and the accuracy of these models is directly related to the effectiveness of the model. In addition, in the case of a special purpose company, whether or not to purchase the property is the main concern. Therefore, whether or not to achieve a certain level of return is enough to make a decision. For the dependent variable, we constructed and compared the predictive model by calculating the dependent variable by adjusting the numerical value to ascertain whether 12%, which is the standard rate of return used in the industry, is a meaningful reference value. As a result, it was found that the hit ratio average of the predictive model constructed using the dependent variable calculated by the 12% standard rate of return was the best at 64.60%. In order to propose an optimal prediction model based on the determined dependent variables and 7 independent variables, we construct a prediction model by applying the five methodologies of discriminant analysis, logistic regression analysis, decision tree, artificial neural network, and genetic algorithm linear model we tried to compare them. To do this, 10 sets of training data and testing data were extracted using 10 fold validation method. After building the model using this data, the hit ratio of each set was averaged and the performance was compared. As a result, the hit ratio average of prediction models constructed by using discriminant analysis, logistic regression model, decision tree, artificial neural network, and genetic algorithm linear model were 64.40%, 65.12%, 63.54%, 67.40%, and 60.51%, respectively. It was confirmed that the model using the artificial neural network is the best. Through this study, it is proved that it is effective to utilize 7 independent variables and artificial neural network prediction model in the future NPL market. The proposed model predicts that the 12% return of new things will be achieved beforehand, which will help the special purpose companies make investment decisions. Furthermore, we anticipate that the NPL market will be liquidated as the transaction proceeds at an appropriate price.

• Journal of Korean Society of Transportation
• /
• v.22 no.3 s.74
• /
• pp.145-157
• /
• 2004

The cycle length design model of the Korean traffic responsive signal control systems is devised to vary a cycle length as a response to changes in traffic demand in real time by utilizing parameters specified by a system operator and such field information as degrees of saturation of through phases. Since no explicit guideline is provided to a system operator, the system tends to include ambiguity in terms of the system optimization. In addition, the cycle lengths produced by the existing model have yet been verified if they are comparable to the ones minimizing delay. This paper presents the studies conducted (1) to find shortcomings embedded in the existing model by comparing the cycle lengths produced by the model against the ones minimizing delay and (2) to propose a new direction to design a cycle length minimizing delay and excluding such operator oriented parameters. It was found from the study that the cycle lengths from the existing model fail to minimize delay and promote intersection operational conditions to be unsatisfied when traffic volume is low, due to the feature of the changed target operational volume-to-capacity ratio embedded in the model. The 64 different neural network based cycle length design models were developed based on simulation data surrogating field data. The CORSIM optimal cycle lengths minimizing delay were found through the COST software developed for the study. COST searches for the CORSIM optimal cycle length minimizing delay with a heuristic searching method, a hybrid genetic algorithm. Among 64 models, the best one producing cycle lengths close enough to the optimal was selected through statistical tests. It was found from the verification test that the best model designs a cycle length as similar pattern to the ones minimizing delay. The cycle lengths from the proposed model are comparable to the ones from TRANSYT-7F.

• The Journal of the Petrological Society of Korea
• /
• v.18 no.3
• /
• pp.223-236
• /
• 2009

Ultramafic xenoliths from southeastern part of Jeju Island can be grouped into two types: Type I and Type II. Type I xenoliths are magnesian and olivine-rich peridotite (mg#=89-91), which are commonly found at the outcrop. Most previous works have been focused on Type I xenoliths. Type II xenoliths, consisting of olivine, orthopyroxene and clinopyroxene with higher Fe and Ti components (mg#=77-83) and lower Mg, Ni, Cr, are reported in this study. They are less common with a more extensive compositional range. The studied Type II xenoliths are wehrlite, olivine-clinopyroxenite, olivine websterite, and websterite. They sometimes show ophitic textures in outcrops indicating cumulate natures. The textural characteristics, such as kink banding and more straight grain boundaries with triple junctions, are interpreted as the result of recrystallization and annealing. Large pyroxene grains have exsolution textures and show almost the same major compositions as small exsolution-free pyroxenes. Although the exsolution texture indicates a previous high-temperature history, all mineral phases are completely reequilibrated to some lower temperature. Orthopyroxenes replacing clinopyroxene margin or olivine indicate an orthopyroxene enrichment event. Mineral phases of Type II are compared with Type I xenoliths, gabbroic xenoliths, and the host basalts. Those from Type II xenoliths show a distinct discontinuity with those from Type I mantle xenoliths, whereas they show a continuous or overlapping relation with those from gabbroic xenoliths and the host basalts. Our petrographic and geochemical results suggest that the studied type II xenoliths appear to be cumulates derived from the host magma-related system, being formed by early fractional crystallization, although these xenoliths may not be directly linked to the host basalt.

• Journal of Intelligence and Information Systems
• /
• v.22 no.1
• /
• pp.139-157
• /
• 2016

Bankruptcy involves considerable costs, so it can have significant effects on a country's economy. Thus, bankruptcy prediction is an important issue. Over the past several decades, many researchers have addressed topics associated with bankruptcy prediction. Early research on bankruptcy prediction employed conventional statistical methods such as univariate analysis, discriminant analysis, multiple regression, and logistic regression. Later on, many studies began utilizing artificial intelligence techniques such as inductive learning, neural networks, and case-based reasoning. Currently, ensemble models are being utilized to enhance the accuracy of bankruptcy prediction. Ensemble classification involves combining multiple classifiers to obtain more accurate predictions than those obtained using individual models. Ensemble learning techniques are known to be very useful for improving the generalization ability of the classifier. Base classifiers in the ensemble must be as accurate and diverse as possible in order to enhance the generalization ability of an ensemble model. Commonly used methods for constructing ensemble classifiers include bagging, boosting, and random subspace. The random subspace method selects a random feature subset for each classifier from the original feature space to diversify the base classifiers of an ensemble. Each ensemble member is trained by a randomly chosen feature subspace from the original feature set, and predictions from each ensemble member are combined by an aggregation method. The k-nearest neighbors (KNN) classifier is robust with respect to variations in the dataset but is very sensitive to changes in the feature space. For this reason, KNN is a good classifier for the random subspace method. The KNN random subspace ensemble model has been shown to be very effective for improving an individual KNN model. The k parameter of KNN base classifiers and selected feature subsets for base classifiers play an important role in determining the performance of the KNN ensemble model. However, few studies have focused on optimizing the k parameter and feature subsets of base classifiers in the ensemble. This study proposed a new ensemble method that improves upon the performance KNN ensemble model by optimizing both k parameters and feature subsets of base classifiers. A genetic algorithm was used to optimize the KNN ensemble model and improve the prediction accuracy of the ensemble model. The proposed model was applied to a bankruptcy prediction problem by using a real dataset from Korean companies. The research data included 1800 externally non-audited firms that filed for bankruptcy (900 cases) or non-bankruptcy (900 cases). Initially, the dataset consisted of 134 financial ratios. Prior to the experiments, 75 financial ratios were selected based on an independent sample t-test of each financial ratio as an input variable and bankruptcy or non-bankruptcy as an output variable. Of these, 24 financial ratios were selected by using a logistic regression backward feature selection method. The complete dataset was separated into two parts: training and validation. The training dataset was further divided into two portions: one for the training model and the other to avoid overfitting. The prediction accuracy against this dataset was used to determine the fitness value in order to avoid overfitting. The validation dataset was used to evaluate the effectiveness of the final model. A 10-fold cross-validation was implemented to compare the performances of the proposed model and other models. To evaluate the effectiveness of the proposed model, the classification accuracy of the proposed model was compared with that of other models. The Q-statistic values and average classification accuracies of base classifiers were investigated. The experimental results showed that the proposed model outperformed other models, such as the single model and random subspace ensemble model.

• Clinical and Experimental Reproductive Medicine
• /
• v.31 no.1
• /
• pp.29-39
• /
• 2004

Objectives: This study was performed to evaluate the laboratory system for successful PGD using fluorescence in situ hybridization (FISH) and the clinical outcome of PGD cycles in five years experiences. Methods: A total of 181 PGD-FISH cycles of 106 couples were performed, and diagnosed chromosome normality in the preimplantation embryos. The laboratory and clinical data were classified by the following optimization steps, and statistically analyzed. Phase I: Blastomere biopsy with two kinds of pipettes, removal of cytoplasmic proteins without treatment of pepsin and culture of biopsied embryos with single medium; Phase II: Blatomere biopsy with single pipette, removal of cytoplasmic proteins with pepsin and culture of biopsied embryos with single medium; Phase III: Blastomere biopsy with single pipette, removal of cytoplasmic proteins with pepsin and culture of biopsied embryos with sequential media. Results: A total of 3, 209 oocytes were collected, and 83.8% (2, 212/2, 640) of fertilization rate was obtained by ICSI procedure. The successful blastomere biopsies were accomplished in 98.6% (2, 043/2, 071) of embryos, and the successful diagnosis rate of FISH was 94.7% (1, 935/ 2, 043) of blastomeres from overall data. Embryo transfers with normal embryos were conducted in 93.9% (170/181) of started cycles. There was no difference in the successful rate of biopsy and diagnosis among Phase I, II and III. However, the pregnancy rate per embryo transfer of Phase III (38.8%, 26/67) was significantly (p<0.05) higher than those of Phase I (13.9%, 5/36) and Phase II (14.9%, 10/67). Conclusions: The laboratory optimization and experience for the PGD with FISH procedure can increase the pregnancy rate to 38.8% in the human IVF-ET program. Our facility of PGD with FISH provides the great possibility to get a normal pregnancy for the concerned couples by chromosomal aberrations.

• Journal of Plant Biotechnology
• /
• v.33 no.1
• /
• pp.33-37
• /
• 2006

Bottle gourd (Lagenaria siceraria Standl.) has been used as a rootstock for the watermelon cultivation because of better growth ability at low temperature and avoidance from contamination of the soil disease. Since the genetic source for the elite rootstock is limited in nature, the genetic engineering method is inevitable to develop new lines especially to obtain the functionally important or multi-disease resistant bottle gourd. Recently, our lab has set up a successful system to transform the bottle gourd. in order to monitor the transformation process, GFP gene is used. Cotyledons of the inbred line 9005, 9006 and G5 were used to induce the shoot under the selection media with MS + 30 g/L sucrose + 3.0 mg/L BAP + 100 mg/L kanamycin + 500 mg/L cefotaxime + 0.5 mg/L $AgNO_3$, pH 5.8. The shoot was developed from the cut side of the explants after 3 weeks on the selection media. The shoot was incubated in the rooting media with 1/2 MS + 30 g/L sucrose + 0.1 mg/L IAA + 50 mg/L kanamycin + 500 mg/L cefotaxime, pH 5.8 and moved to pot for acclimation. Although the shoot development rate was depended on the genotype, the G5 was the best line to be transformed. Monitoring GFP expression from the young shoot under microscope could make the selection much easier to distinguish the transformed shoot from the non-transformed shoots.

• Radiation Oncology Journal
• /
• v.24 no.2
• /
• pp.128-137
• /
• 2006

Purpose: We investigated whether a relationship exists between tumor control dose 50 ($TCD_{50}$) or tumor growth delay (TGD) and radiation induced apoptosis (RIA) in syngeneic murine tumors. Also we investigated the biological markers that can predict radiosensitivity in murine tumor system through analysis of relationship between $TCD_{50}$, TGD, RIA and constitutive expression levels of the genetic products regulating RIA. Materials and Methods: Syngeneic murine tumors such as ovarian adenocarcinoma, mammary carcinoma, squamous cell carcinoma, fibrosarcoma, hepatocarcinoma were used In this study. C3H/HeJ mice were bred and maintained in our specific pathogen free mouse colony and were $8{\sim}12$ weeks old when used for the experiments. The tumors, growing in the right hind legs of mice, were analyzed for $TCD_{50}$, TGD, and RIA at 8 mm in diameter. The tumors were also analyzed for the constitutive expression levels of $p53,\;p21^{WAF1/CIP1},\;BAX,\;Bcl-2,\;Bcl-X_L,\;Bcl-X_S$, and p34. Correlation analysis was peformed whether the level of RIA were correlated with $TCD_{50}$ or TGD, and the constitutive expression levels of genetic products regulating RIA were correlated with $TCD_{50}$, TGD, RIA. Results: The level of RIA showed a significant positive correlation (R=0.922, p=0.026) with TGD, and showed a trend to correlation (R=-0.848), marginally significant correlation with $TCD_{50}$ (p=0.070). It indicates that tumors that respond to radiation with high percentage of apoptosis were more radiosensitive. The constitutive expression levels of $p21^{WAF1/CIP1}$ and 34 showed a significant correlation either with $TCD_{50}$ (R=0.893, p=0.041 and R=0.904, p=0.035) or with TGD (R=-0.922, p=0.026 and R=-0.890 p=0.043). The tumors with high constitutive expression levels of $p21^{WAF1/CIP1}$ or p34 were less radiosensitive than those with low expression. Conclusion: Radiosensitivity may be predicted with the level of RIA in murine tumors. The constitutive expression levels of $p21^{WAF1/CIP1}$ or p34 can be used as biological markers which predict the radiosensitivity.

• Research in Plant Disease
• /
• v.19 no.4
• /
• pp.254-258
• /
• 2013

Fusaric acid (FA) is a mycotoxin produced by Fusarium species. Its toxicity is relatively low but often associated with other mycotoxins, thus enhancing total toxicity. To date, biosynthetic genes or enzymes for FA have not been identified in F. oxysporum. In order to explore the genetic element(s) for FA biosynthesis, restriction enzyme mediated integration (REMI) procedure as an insertional mutagenesis was employed using FA producing-F. oxysporum strains. Genetic transformation of two F. oxysporum strains by REMI yielded more than 7,100 transformants with efficiency of average 3.2 transformants/${\mu}g$ DNA. To develop a screening system using phytotoxicity of FA, eleven various grains and vegetable seeds were tested for germination in cultures containing FA: Kimchi cabbage seed was selected as the most sensitive host. Screening for FA non-producer of F. oxysporum was done by growing each fungal REMI transformant in Czapek-Dox broth for 3 weeks at $25^{\circ}C$ then observing if the Kimchi cabbage seeds germinated in the culture filtrate. Of more than 5,000 REMI transformants screened, fifty-three made the seeds germinated, indicating that they produced little or fewer FA. Among them, twenty-six were analyzed for FA production by HPLC and two turned out to produce less than 1% of FA produced by a wild type strain. Sequencing of genomic DNA regions (252 bp) flanking the vector insertion site revealed an uncharacterized genomic region homologous (93%) to the F. fujikuroi genome. Further study is necessary to determine if the vector insertion sites in FA-deficient mutants are associated with FA production.

• Journal of Intelligence and Information Systems
• /
• v.17 no.4
• /
• pp.241-254
• /
• 2011

Financial time-series forecasting is one of the most important issues because it is essential for the risk management of financial institutions. Therefore, researchers have tried to forecast financial time-series using various data mining techniques such as regression, artificial neural networks, decision trees, k-nearest neighbor etc. Recently, support vector machines (SVMs) are popularly applied to this research area because they have advantages that they don't require huge training data and have low possibility of overfitting. However, a user must determine several design factors by heuristics in order to use SVM. For example, the selection of appropriate kernel function and its parameters and proper feature subset selection are major design factors of SVM. Other than these factors, the proper selection of instance subset may also improve the forecasting performance of SVM by eliminating irrelevant and distorting training instances. Nonetheless, there have been few studies that have applied instance selection to SVM, especially in the domain of stock market prediction. Instance selection tries to choose proper instance subsets from original training data. It may be considered as a method of knowledge refinement and it maintains the instance-base. This study proposes the novel instance selection algorithm for SVMs. The proposed technique in this study uses genetic algorithm (GA) to optimize instance selection process with parameter optimization simultaneously. We call the model as ISVM (SVM with Instance selection) in this study. Experiments on stock market data are implemented using ISVM. In this study, the GA searches for optimal or near-optimal values of kernel parameters and relevant instances for SVMs. This study needs two sets of parameters in chromosomes in GA setting : The codes for kernel parameters and for instance selection. For the controlling parameters of the GA search, the population size is set at 50 organisms and the value of the crossover rate is set at 0.7 while the mutation rate is 0.1. As the stopping condition, 50 generations are permitted. The application data used in this study consists of technical indicators and the direction of change in the daily Korea stock price index (KOSPI). The total number of samples is 2218 trading days. We separate the whole data into three subsets as training, test, hold-out data set. The number of data in each subset is 1056, 581, 581 respectively. This study compares ISVM to several comparative models including logistic regression (logit), backpropagation neural networks (ANN), nearest neighbor (1-NN), conventional SVM (SVM) and SVM with the optimized parameters (PSVM). In especial, PSVM uses optimized kernel parameters by the genetic algorithm. The experimental results show that ISVM outperforms 1-NN by 15.32%, ANN by 6.89%, Logit and SVM by 5.34%, and PSVM by 4.82% for the holdout data. For ISVM, only 556 data from 1056 original training data are used to produce the result. In addition, the two-sample test for proportions is used to examine whether ISVM significantly outperforms other comparative models. The results indicate that ISVM outperforms ANN and 1-NN at the 1% statistical significance level. In addition, ISVM performs better than Logit, SVM and PSVM at the 5% statistical significance level.

• Proceedings of the Korean Society of Crop Science Conference
• /
• 2017.06a
• /
• pp.8-9
• /
• 2017

Soybean yield has been low and unstable in Japan and other areas in East Asia, despite long history of cultivation. This is contrasting with consistent increase of yield in North and South America. This presentation tries to describe perspective of breaking stagnation of soybean yield in East Asia, considering the factors of the different yields between regions. Large amount of rainfall with occasional dry-spell in the summer is a nature of monsoon climate and as frequently stated excess water is the factor of low and unstable soybean yield. For example, there exists a great deal of field-to-field variation in yield of 'Tanbaguro' soybean, which is reputed for high market value and thus cultivated intensively and this results in low average yield. According to our field survey, a major portion of yield variation occurs in early growth period. Soybean production on drained paddy fields is also vulnerable to drought stress after flowering. An analysis at the above study site demonstrated a substantial field-to-field variation of canopy transpiration activity in the mid-summer, but the variation of pod-set was not as large as that of early growth. As frequently mentioned by the contest winners of good practice farming, avoidance of excess water problem in the early growth period is of greatest importance. A series of technological development took place in Japan in crop management for stable crop establishment and growth, that includes seed-bed preparation with ridge and/or chisel ploughing, adjustment of seed moisture content, seed treatment with mancozeb+metalaxyl and the water table control system, FOEAS. A unique success is seen in the tidal swamp area in South Sumatra with the Saturated Soil Culture (SSC), which is for managing acidity problem of pyrite soils. In 2016, an average yield of $2.4tha^{-1}$ was recorded for a 450 ha area with SSC (Ghulamahdi 2017, personal communication). This is a sort of raised bed culture and thus the moisture condition is kept markedly stable during growth period. For genetic control, too, many attempts are on-going for better emergence and plant growth after emergence under excess water. There seems to exist two aspects of excess water resistance, one related to phytophthora resistance and the other with better growth under excess water. The improvement for the latter is particularly challenging and genomic approach is expected to be effectively utilized. The crop model simulation would estimate/evaluate the impact of environmental and genetic factors. But comprehensive crop models for soybean are mainly for cultivations on upland fields and crop response to excess water is not fully accounted for. A soybean model for production on drained paddy fields under monsoon climate is demanded to coordinate technological development under changing climate. We recently recognized that the yield potential of recent US cultivars is greater than that of Japanese cultivars and this also may be responsible for different yield trends. Cultivar comparisons proved that higher yields are associated with greater biomass production specifically during early seed filling, in which high and well sustained activity of leaf gas exchange is related. In fact, the leaf stomatal conductance is considered to have been improved during last a couple of decades in the USA through selections for high yield in several crop species. It is suspected that priority to product quality of soybean as food crop, especially large seed size in Japan, did not allow efficient improvement of productivity. We also recently found a substantial variation of yielding performance under an environment of Indonesia among divergent cultivars from tropical and temperate regions through in a part biomass productivity. Gas exchange activity again seems to be involved. Unlike in North America where transpiration adjustment is considered necessary to avoid terminal drought, under the monsoon climate with wet summer plants with higher activity of gas exchange than current level might be advantageous. In order to explore higher or better-adjusted canopy function, the methodological development is demanded for canopy-level evaluation of transpiration activity. The stagnation of soybean yield would be broken through controlling variable water environment and breeding efforts to improve the quality-oriented cultivars for stable and high yield.