• Journal of the Korean Data and Information Science Society
• /
• v.24 no.6
• /
• pp.1113-1125
• /
• 2013

Corporate bankruptcy prediction has been an important topic in the accounting and finance field for a long time. Several data mining techniques have been used for bankruptcy prediction. However, there are many limits for application to real classification problem with a single model. This study proposes ensemble SVM (support vector machine) model which assembles different SVM models with each different kernel functions. Our ensemble model is made and evaluated by v-fold cross-validation approach. The k top performing models are recruited into the ensemble. The classification is then carried out using the majority voting opinion of the ensemble. In this paper, we investigate the performance of ensemble SVM classifier in terms of accuracy, error rate, sensitivity, specificity, ROC curve, and AUC to compare with single SVM classifiers based on financial ratios dataset and simulation dataset. The results confirmed the advantages of our method: It is robust while providing good performance.

• Journal of KIISE:Information Networking
• /
• v.30 no.4
• /
• pp.520-534
• /
• 2003

In this paper, a performance evaluation model of the Fat-tree Network with the multiple-buffered crossbar switches is proposed and examined. Buffered switch technique is well known to solve the data collision problem of the switch network. The proposed evaluation model is developed by investigating the transfer patterns of data packets in a switch with output-buffers. Two important parameters of the network performance, throughput and delay, are then evaluated. The proposed model takes simple and primitive switch networks, i.e., no flow control and drop packet, to demonstrate analysis procedures clearly. It, however, can not only be applied to any other complicate modern switch networks that have intelligent flow control but also estimate the performance of any size networks with multiple-buffered switches. To validate the proposed analysis model, the simulation is carried out on the various sizes of Fat-tree networks that uses the multiple buffered crossbar switches. Less than 2% differences between analysis and simulation results are observed.

• Journal of the Korea Concrete Institute
• /
• v.22 no.3
• /
• pp.409-417
• /
• 2010

This paper discusses a series of experiments including material improvement in order to ensure quality of grouting for the post-tensioned structure. In prestressed concrete, grouting refers to the construction procedure of filling empty space of duct enclosing with strands using cementitious material, To date, adequate quality control of the grouting has not been established in Korea because the relationship between the grouting and durability of post-tensioned structure is not well-recognized. The Korean standard does not consider the important material characteristic, wick effect, which is caused by strands in the ducts and current standard testing method unlikely quantify reasonable material segregation. As a result, the grout material, which meets the current material standards, may exhibit excessive bleeding water or shrinkage during construction. In this study, international codes and standards related to grouting were surveyed. The ratio of constituents and novel admixtures were suggested to meet equivalently with these standards. Performance of this enhanced grout was compared to common domestic grout using the international standard testing method. A series of mock-up specimens considering geometry of PC beam was constructed and grout flow pattern was observed as the grout was injected. It was observed that the grouting performance was highly influenced by material properties and filling characteristic can be varied depending on geometry of ducts.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.10 no.5
• /
• pp.179-186
• /
• 2006

In this study, Multi-Layer Perceptron(MLP) among models of Artificial Neural Network(ANN) is used for the development of a model that evaluates the bending capacities of reinforced concrete beams strengthened by FRP Rebar. And the data of the existing researches are used for materials of ANN model. As the independent variables of input layer, main components of bending capacities, width, effective depth, compressive strength, reinforcing ratio of FRP, balanced steel ratio of FRP are used. And the moment performance measured in the experiment is used as the dependent variable of output layer. The developed model of ANN could be applied by GFRP, CFRP and AFRP Rebar and the model is verified by using the documents of other previous researchers. As the result of the ANN model presumption, comparatively precise presumption values are achieved to presume its bending capacities at the model of ANN(0.05), while observing remarkable errors in the model of ANN(0.1). From the verification of the ANN model, it is identified that the presumption values comparatively correspond to the given data ones of the experiment. In addition, from the Sensitivity Analysis of evaluation variables of bending performance, effective depth has the highest influence, followed by steel ratio of FRP, balanced steel ratio, compressive strength and width in order.

• Journal of the Korea Society of Computer and Information
• /
• v.16 no.4
• /
• pp.29-42
• /
• 2011

This study deals with the assessment indices about greening performance of datacenter. To do this, we survey existing standards and guidelines about datacenter. Those are used constructing new model of datacenter regarding greening performance. In this model, the relationship between major components will be represented by job-energy-thermal notation. Items, criteria and methods are created in order to complete the assessment indices. The existing assessment indices of the datacenter level and equipment level are investigated. Parameters, the degree of virtualization and energy efficiency also affects the performance of the green performance. Finally, to determine the improvements, the proposed indices is compared with the existing assessment tools.

• Journal of Korean Society of Transportation
• /
• v.28 no.6
• /
• pp.75-84
• /
• 2010

Controlled traffic intersection is critical point in terms of transportation network performance, where the most of traffic congestion arises. One of the most important and favorable measure of effectiveness in the signal controlled intersection is approach delay. Although lots of efforts to develop traffic delay estimation models have been made throughout the years, most of them were focusing on homogeneous traffic flow. The purpose of this research is to develop a traffic delay estimation model for traffic flow mixed with bus based on the horizontal shockwave theory. Traffic simulation is performed to test the adaptation level of the model in generic environment. The result shows that the delay increases with increasing bus traffic. Overall model accuracy comparing simulation result is acceptable, that shows the error range around 10 percent.

• Journal of Korea Society of Industrial Information Systems
• /
• v.6 no.1
• /
• pp.99-108
• /
• 2001

사용자 만족(User satisfaction)은 Bailey와 Pearson(1983) 이후, 정보시스템 분야의 연구에서 시스템의 성능(Performance)을 대변하는 성과변수로서 가장 널리 사용되어져 왔다.[Delone & McLean, 1992]. 그러나 사용자 만족은 실제 시스템의 객관적인 성능이나 품질이 아닌 사용자에 의해 지각된(Perceived) 만족이므로, 인간이 인지과정에서 받는 다른 영향들로 인하여 지각된 성과는 실제의 시스템 성능과 다를 수 있다.[Gatian 1994; Szajna & Scamell, 1993. 따라서 만일 사용자가 지각한 성능이 객관적인 실제의 시스템 성능과 반드시 일치하지 않는다면, 그 원인은 무엇이며, 사용자가 시스템을 평가하는 과정에서 어떤 요인들이 작용하여 이러한 왜곡된 결과를 가져오는지, 사용자 만족이 결정되는 인지적 과정을 규명할 필요가 있다. 이러한 의미에서 본 연구는 마케팅 분야에서 일반상품에 대한 소비자의 만족 결정에 영향을 주는 변수와 그 결정과정을 연구한 소비자 만족/불만족(Consumer Satisfaction / Dissatisfaction)의 연구이론을 도입하여, 정보시스템이란 상품을 대상으로 사용자가 만족을 느끼게 되는 과정을 인지적(Cognitive) 관점에서 규명하고, 만족결정에 영향을 미치는 변수들을 찾아낸 후 소비자 만족을 결정하는 모형을 연구하고자 하였다. 8개 기업의 정보시스템 사용자부터 데이터를 수집하여 LISREL을 이용하여 사용자 만족 연구 모형을 검증하였다. 분석 결과, 결정모형은 유의하였으며 정보시스템의 사용자 만족에 영향을 미치는 변수로는 시스템의 성능 뿐 아니라 기대불일치가 함께 영향을 미치는 것으로 나타나 시스템 성능의 향상과 함께 사용자들의 기대수준 관리에 관심을 기울일 필요가 있는 것으로 나타났다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.351-351
• /
• 2023

강우 자료는 수문 모델링에서 중요한 입력 요소 중 하나이다. 강우의 공간적 가변성은 모델링 불확실성의 중요한 원인으로 알려져 있다. 강우 관측자료는 많은 경우 유역을 대표하는 평균 면적강수량 (Mean Areal Precipitation, MAP)을 계산하여 수문모형에 입력된다. 선행 연구에서는MAP 예측 결과의 신뢰도를 개선하기 위하여 다양한 보간 방법이 개발되었다. 하지만, 강우특성의 동질성를 고려한 대표 기상 관측소 선정이 MAP 예측과 유출량 모의 결과에 미치는 연구는 아직 미흡한 실정이다. 본 연구에서는 유역의 MAP 예측에 있어 강우특성의 동실성을 고려한 강우 관측소 선정이 수문 모델링 성능 개선에 미치는 영향을 평가하고자 한다. 본 연구에서는 종관 기상관측(ASOS) 74개 지점과 방재기상관측(AWS) 400여개 지점에서 2003~2022년 기간에 대한 일강수량 자료를 수집하였고 강우특성이 동질한 지역을 구분하였다. 또한, 강우특성 동질성의 고려 유무에 따른 MAP를 계산하였다. 이후, 5개의 매개변수로 이루어진 개념적 강우-유출 모형FPHM을 사용하여 우리나라 전역 41개 유역을 대상으로 MAP 계산 결과가 모형 성능에 미치는 민감도를 조사하였다. 분석 결과, 강우특성의 동질성을 고려한 강우 관측소의 선택은 MAP 보간 방법 이상으로 중요한 요소임을 확인할 수 있었다.

• Proceedings of the KAIS Fall Conference
• /
• 2009.12a
• /
• pp.29-32
• /
• 2009

본 연구는 도로교 설계 기준 바탕으로 H-형강을 사용하여 20~30m 사이의 지간장을 가진 중 소규모 교량에 적용이 가능한 초간편 교량의 모형실험 결과와 유한요소해석프로그램(ABAQUS)을 사용한 해석 결과와 비교 분석하여 초간편 교량의 성능을 평가한 것이다. 일반적으로 우리나라의 교량들은 시공과정에서 여러 단계를 거쳐 시공한다. 시공단계가 복잡할수록 공사기간은 늘어나기 때문에 기상 현상에 의한 파괴와 교량의 낙후로 인해 유지, 보수, 교체 또는 교량확장을 해야 하는 경우 교통 혼잡과 경제적 손실을 줄 수 있다. 따라서 본 연구는 이와 같은 경제적 손실을 줄일 수 있는 초간편 H형강 교량의 연구의 일환으로 기존의 연구 결과를 모형실험 수행 결과와 비교하여 최적성능을 평가하기 위해 실시하였다. 사용된 실험체는 실험장소인 건설기술연구소의 구조실험동 장소 여건에 따라 10m 안팎의 경간을 갖는 교량으로 제작하였다. 실험체의 설계 제작 과정에 대해 검토한 후 범용구조해석 프로그램을 이용하여 동일한 조건을 적용한 결과를 비교 후 최종적으로 초간편 교량의 성능에 대한 평가를 실시하여 김재흥 등(2009)의 의해 제안된 하중분배계수 값이 적용가능한지 판단하였고, 실험을 통해 응력 분포와 극한하중에 대해 검토하였다.

• Fire Science and Engineering
• /
• v.28 no.2
• /
• pp.20-25
• /
• 2014

In this study, the extinguishing performance evaluation of wetting agent for wood crib was conducted by using a scale model equipment that we designed. To confirm the optimal conditions of the experiment, a test was changed amount of fire extinguishing water and the number of timber. As a result, the discrimination of the fire extinguishing performance was seen only when 20 pieces of wood and the extinguishing water of 100 mL were used. After evaluating the extinguishing performance of domestic and foreign wetting extinguishing agents under these conditions, a reignition was occurred in only when we used water. In addition, the discrimination of extinguishing performance was seen through the temperature distribution according to the time of watering. Based on the results of this experiment, this study is expected to be able to use as a basis on presenting a method of optimized performance evaluation of wetting extinguishing agent.