• Journal of Intelligence and Information Systems
• /
• v.20 no.4
• /
• pp.121-139
• /
• 2014

Predicting corporate failure has been an important topic in accounting and finance. The costs associated with bankruptcy are high, so the accuracy of bankruptcy prediction is greatly important for financial institutions. Lots of researchers have dealt with the topic associated with bankruptcy prediction in the past three decades. The current research attempts to use ensemble models for improving the performance of bankruptcy prediction. Ensemble classification is to combine individually trained classifiers in order to gain more accurate prediction than individual models. Ensemble techniques are shown to be very useful for improving the generalization ability of the classifier. Bagging is the most commonly used methods for constructing ensemble classifiers. In bagging, the different training data subsets are randomly drawn with replacement from the original training dataset. Base classifiers are trained on the different bootstrap samples. Instance selection is to select critical instances while deleting and removing irrelevant and harmful instances from the original set. Instance selection and bagging are quite well known in data mining. However, few studies have dealt with the integration of instance selection and bagging. This study proposes an improved bagging ensemble based on instance selection using genetic algorithms (GA) for improving the performance of SVM. GA is an efficient optimization procedure based on the theory of natural selection and evolution. GA uses the idea of survival of the fittest by progressively accepting better solutions to the problems. GA searches by maintaining a population of solutions from which better solutions are created rather than making incremental changes to a single solution to the problem. The initial solution population is generated randomly and evolves into the next generation by genetic operators such as selection, crossover and mutation. The solutions coded by strings are evaluated by the fitness function. The proposed model consists of two phases: GA based Instance Selection and Instance based Bagging. In the first phase, GA is used to select optimal instance subset that is used as input data of bagging model. In this study, the chromosome is encoded as a form of binary string for the instance subset. In this phase, the population size was set to 100 while maximum number of generations was set to 150. We set the crossover rate and mutation rate to 0.7 and 0.1 respectively. We used the prediction accuracy of model as the fitness function of GA. SVM model is trained on training data set using the selected instance subset. The prediction accuracy of SVM model over test data set is used as fitness value in order to avoid overfitting. In the second phase, we used the optimal instance subset selected in the first phase as input data of bagging model. We used SVM model as base classifier for bagging ensemble. The majority voting scheme was used as a combining method in this study. This study applies the proposed model to the bankruptcy prediction problem using a real data set from Korean companies. The research data used in this study contains 1832 externally non-audited firms which filed for bankruptcy (916 cases) and non-bankruptcy (916 cases). Financial ratios categorized as stability, profitability, growth, activity and cash flow were investigated through literature review and basic statistical methods and we selected 8 financial ratios as the final input variables. We separated the whole data into three subsets as training, test and validation data set. In this study, we compared the proposed model with several comparative models including the simple individual SVM model, the simple bagging model and the instance selection based SVM model. The McNemar tests were used to examine whether the proposed model significantly outperforms the other models. The experimental results show that the proposed model outperforms the other models.

• Journal of Soil and Groundwater Environment
• /
• v.11 no.5
• /
• pp.9-19
• /
• 2006

The accurate estimation of groundwater recharge is important for the proper management of groundwater systems. The widely used techniques of groundwater recharge estimation include water table fluctuation method, baseflow separation method, and annual water balance method. However, these methods can not represent the temporal-spatial variability of recharge resulting from climatic condition, land use, soil storage and hydrogeological heterogeneity because the methods are all based on the lumped concept and local scale problems. Therefore, the objective of this paper is to present an effective method for estimating groundwater recharge with spatial-temporal variability using the SWAT model which can represent the heterogeneity of the watershed. The SWAT model can simulate daily surface runoff, evapotranspiration, soil storage, recharge, and groundwater flow within the watershed. The model was applied to the Musimcheon watershed located in the upstream of Mihocheon watershed. Hydrological components were determined during the period from 2001 to 2004, and the validity of the results was tested by comparing the estimated runoff with the observed runoff at the outlet of the catchment. The results of temporal and spatial variations of groundwater recharge were presented here. This study suggests that variations in recharge can be significantly affected by subbasin slope as well as land use.

• Journal of Korea Water Resources Association
• /
• v.52 no.11
• /
• pp.931-940
• /
• 2019

In hydrologic models, parameters are mainly used to reflect hydrologic elements or to supplement the simplified models. In this process, the proper selection of the parameters in the model can reduce the uncertainty. Accordingly, this study attempted to quantify the uncertainty of SWAT parameters using the General Likelihood Uncertainty Estimation (GLUE). Uncertainty analysis on SWAT parameters was conducted by using the formal and informal likelihood measures. The Lognormal function and Nash-Sutcliffe Efficiency (NSE) were used for formal and informal likelihood, respectively. Subjective factors are included in the selection of the likelihood function and the threshold, but the behavioral models were created by selecting top 30% lognormal for formal likelihood and NSE above 0.5 for informal likelihood. Despite the subjectivity in the selection of the likelihood and the threshold, there was a small difference between the formal and informal likelihoods. In addition, among the SWAT parameters, ALPHA_BF which reflects baseflow characteristics is the most sensitive. Based on this study, if the range of SWAT model parameters satisfying a certain threshold for each watershed is classified, it is expected that users will have more practical or academic access to the SWAT model.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.2
• /
• pp.404-411
• /
• 2019

In this study, the HEC-HMS was applied to determine rainfall-runoff processes for the Gokgyuchun basin. Several sub-basins have large-scale reservoirs for agricultural needs and they store large amounts of initial runoff. Three infiltration methods were implemented to reflect the effect of initial loss by reservoirs: 'SCS-CN'(Scheme I), 'SCS-CN' with simple surface method(Scheme II), and 'Initial and Constant rate'(Scheme III). Modeling processes include incorporating three different methods for loss due to infiltration, Clark's UH model for transformation, exponential recession model for baseflow, and Muskingum model for channel routing. The parameters were calibrated using an optimization technique with trial and error method. Performance measures, such as NSE, RAR, and PBIAS, were adopted to aid in the calibration processes. The model performance for those methods was evaluated at Gangcheong station, which is the outlet of study site. Good accuracy in predicting runoff volume and peak flow, and peak time was obtained using the Scheme II and III, considering the initial loss, whereas Scheme I showed low reliability for storms. Scheme III did not show good matches between observed and simulated values for storms with multi peaks. Conclusively, Scheme II provided better results for both single and multi-peak storms. The results of this study can provide a useful tool for decision makers to determine master plans for regional flood control management.

• Journal of Internet Computing and Services
• /
• v.15 no.1
• /
• pp.143-156
• /
• 2014

The smart education has the goal of enhancing the capability of learners in the 21st century and especially address the improvement of the problem solving capability. This smart education based on the growth of smart devices and the effect of dramatical spread requires the ability of problem solving using the smart technology in accordance with time change. As the problem solving learning is a model used mainly for improving the capability of problem solving, this study develops the problem solving learning model focusing on the teaching-learning activity using the smart devices and also applies this model to the school field. As a result, the favorable response that using the smart devices is effective to the problem solving can be obtained. This study can contribute to achieve the goal of the smart education, and later can be effective to the successful smart education in the school field.

• Korean Journal of Cognitive Science
• /
• v.14 no.1
• /
• pp.41-50
• /
• 2003

We tested the capability of the pleasure and arousal dimension-based facial expression synthesis system and proposed the most effective interface for it. First, we tried to confirm the adequateness of the dimensional model as a basic structure of the internal states for the system. Fer it, subjects compared the 17 facial expressions on the two axes. The results validated the fundamental hypothesis of the system. Second, we chose 21 representative expressions from the system to test its performance and had subjects rate their similarities. We analyzed these data using multidimensional scaling methods and these results verified the system's reliability. Third, we compared the efficiencies of two interfaces -coordinate values and slide bars- to find the most suitable interface for the system. Subjects synthesise 25 facial expressions with each interface of it. The results showed that the visualization of two dimensional values into Cartesian coordinate is more stable as an input display of facial expression synthesis system based on dimensions.

• Journal of Korea Water Resources Association
• /
• v.38 no.6 s.155
• /
• pp.449-460
• /
• 2005

In this study, a WEP (Water and Energy transfer Processes) model was used to simulate the water cycle of the Dorimcheon catchment which suffers from the distorted water cycle as a typical urban catchment. Two different land uses in the past (i.e. 1975) and at present (i.e. 2000) were incorporated into the simulation to investigate the runoff characteristics resulting from the increase of the impervious ratio due to urbanization. The simulation results show that the concentration time is decreased and the peak discharge and the total runoff are increased by urbanization while the infiltration and baseflow are reduced. In addition, the effects of infiltration trenches and permeable pavements were also simulated to search for alternatives that can restore the distorted water cycle. The simulation results prove that the installation of both alternatives can restore the runoff characteristics to that prior to urbanization.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2017.05a
• /
• pp.74-74
• /
• 2017

자연하천에서 수자원의 원활하고 안전한 관리에 있어서 오염물의 혼합 거동에 대한 이해는 매우 중요하다. 대부분의 자연하천의 경우 만곡부 및 합류부와 같은 복잡한 지형을 갖고 있으며 이러한 경우 하천의 흐름이 복잡한 형태를 갖게 된다. 특히 수생태계에 많은 영향을 미치는 하폐수 처리장 처리수는 대부분 1차적으로 지류로 방류되어 이후 본류로 지속적으로 유입되게 된다. 이러한 오염물질이 지류로부터 본류로 혼합되는 합류부 구간의 경우 일반적인 1차원 혼합이 아닌 횡방향을 포함하는 2차원적인 혼합 거동에 대한 분석이 필요하다. 본 연구에서는 금호강과 진천천이 좌안으로부터 오염물질이 지속적으로 유입되는 낙동강 중류구간 합류부에서의 혼합 구간의 연구를 위하여 횡분산계수 산정을 위하여 전기전도도(electrical conductivity: EC)를 이용한 농도 추적 실험을 수행하였다. 낙동강 본류에서 정해진 측선을 따라 센서가 설치된 보트를 이용하여 실시간으로 농도, 수리량 데이터를 GPS 위치 데이터와 함께 취득하였다. 또한 실험으로부터 취득한 자료를 바탕으로 2차원 이송-확산 혼합 거동 모델인 CTM-2D 수치모형을 이용하여 모의하였다. 실험 수행 결과, 지류인 금호강과 진천천의 EC 농도가 합류 전 낙동강 본류의 EC 기저농도 보다 더 높은 값을 나타내었다. 지류의 유입으로 인하여 본류 좌안 쪽에서 전기전도도의 값의 상승을 확인할 수 있었으며 하류로 이동할수록 불균등했던 전기전도도의 분포가 횡방향 혼합을 통하여 점점 균등한 분포로 전환되는 것으로 나타났다. 또한 2차원 혼합 거동 분석에 필요한 횡 분산계수 산정을 위해 모멘트법, 해석해를 이용한 추적법, 수치모형을 통한 역산법을 통해 산정하여 결과를 비교하였다. 그 결과 모멘트법의 경우 다른 방법들에 비하여 전반적으로 과소 산정하는 경향을 나타내었다.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.1
• /
• pp.119-128
• /
• 2023

This study was conducted to present an alternative instructional model through natural experience activities by developing a natural experience activity program that can learn and feel how to recognize and act on nature based on flow learning. In order to achieve the purpose of the study, a nature experience program, which consists of four stages of meeting nature, exploring nature, playing with nature and sharing emotions, was developed based on the main procedures of each stage of the ADDlE instructional design model. Through the research process, activities and precautions for each stage of the nature experience activity program were presented, and major educational implications were discussed based on the developed program. The nature experience program developed through the study can provide teachers with a basic direction for nature experience activities along with changing their perception of how to do nature experience activities, and infants are expected to become learners who freely feel, experience nature and make up their own knowledge through the nature experience program.

• Korean Journal of Agricultural Science
• /
• v.10 no.2
• /
• pp.324-333
• /
• 1983

Modeling of longterm runoff is theoritically based on waterbalance analysis. Simplified equation of water balance with rainfall, evapotranspiration and soil moisture storage could be formulated into regression model with variables of rainfall, pan evaporation and previous-month streamflow. The hydrologic response of water shed could be represented lumpedly, qualitatively and deductively by regression coefficients of water-balance regression model. Characteristics of regression modeling of water-balance were summarized as follows; 1. Regression coefficient $b_1$ represents the rate of direct runoff component of precipitation. The bigger the drainage area, the less $b_1$ value. This means that there are more losses of interception, surface detension and transmission in the downstream watershed. 2. Regression coefficient $b_2$ represents the rate of baseflow due to changes of soil moisture storage. The bigger the drainage area and the milder the watershed slope, the bigger b, value. This means that there are more storage capacity of watershed in mild downstream watershed. 3. Regression coefficient $b_3$ represents the rate of watershed evaporation. This depends on the s oil type, soil coverage and soil moisture status. The bigger the drainage area, the bigger $b_3$ value. This means that there are more watershed evaporation loss since more storage of surface and subsurface water would be in down stream watershed. 4. It was possible to explain the seasonal variation of streamflow reasonably through regress ion coefficients. 5. Percentages of beta coefficients what is a relative measure of the importance of rainfall, evaporation and soil moisture storage to month streamflow are approximately 89%, 9% and 11% respectively.