• The Korean Journal of Applied Statistics
• /
• v.24 no.3
• /
• pp.477-484
• /
• 2011

When an event of target variable is rare, a widespread strategy is to build a model on the sample that disproportionally over-represents the events, that is over-sampled. Using the data over-sampled from the original data set, the predicted values would be biased; however, it can be easily corrected to represent the population. In this study, we investigate into the relationship between the proportion of rare event on a data-mart and the model performance using real world data of a Korean credit card company. Also, we use the methods for adjusting of posterior probability for over-sampled data of the offset method and the weighted method. Finally, we compare the performance of the methods using real data sets.

• The Korean Journal of Applied Statistics
• /
• v.31 no.5
• /
• pp.621-636
• /
• 2018

Research on the application of informative sampling technique has been conducted in order to reduce the influence of non-response. Chung and Shin (Korean Journal of Applied Statistics, 30, 993-1004, 2017) showed that the estimation accuracy improved when using exponential response rate information for the parameter estimation if the distribution of errors included in the super population model follows normal distribution. However this method divides the stratum into equally spaced substrata to obtain the sample weight of the informative sampling technique and shows that the accuracy of the estimation improves as the number of substrata increases. In this study, with the given number of total sample size, the optimal substratum boundary points are calculated using equal space, quantile, and LH algorithm; consequently, the results using those methods are compared through simulation. We also studied the criteria to determine the number of substrata and substratum boundaries that can be used in practice with various types of auxiliary variable distributions.

• Journal of Practical Engineering Education
• /
• v.13 no.2
• /
• pp.393-403
• /
• 2021

The purpose of this study is to propose a method of designing a systematic panel survey at the institutional level to lay the foundation for data-based decision-making using vocational training teachers and instructors as the population. In this study, the target population and sampling frame, which are the main elements necessary for planning a panel survey, are proposed. Also based on expert advice and empirical data analysis, the sampling unit and sampling method taking into account the outer and inner variables are presented, comprehensively considering the representativeness of data, the efficiency and sustainability of data collection. As a result of the study, with the unit of the panel as a vocational training institution, a two-stage stratified proportional sampling plan is proposed so that the institution selected as the panel and the vocational training teachers and instructors belonging to the institution can participate in the survey. Based on this, implications for the panel survey sample design are presented.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2008.11a
• /
• pp.219-222
• /
• 2008

본 논문은 공간영역 상에서 다양하게 변형된 복제 영상과 원본 영상간의 통계적 특성을 이용하여 그 유사도를 측정하고 복제 여부를 판단하는 계층적 구조의 불법 비디오 감식 방법을 제안한다. 영상의 대표적 특성인 명암도에 따라 순위를 매김으로써 공간적 변형에 영향을 받지 않도록 하였으며, 데이터베이스에 저장된 방대한 양의 영상들에 대한 검색 시간과 계산량을 줄이기 위해 순위 표본 프레임을 이용하여 유사한 후보 영상군을 추출한다. 이러한 후보 영상군을 대상으로 일반적인 불법복제 비디오의 형태를 감안하여 각 프레임의 중앙 영역을 포함하여 통계 검정을 이용함으로써 복제 여부를 판단한다. 실험 결과 제안하는 방법은 기존 방법과 유사한 정확도를 보이며 동시에 선택된 순위 표본 프레임 수는 약 50% 가량 적게 추출되어 검색 시간과 계산량이 감소하였다. 또한 영상의 화질 열화, 대비 변형, 확대 및 축소, letterbox 등 다양한 공간적 변형에도 포괄적으로 복제 여부를 판단할 수 있음을 실험을 통해 확인하였다.

• Communications of Mathematical Education
• /
• v.19 no.1 s.21
• /
• pp.151-165
• /
• 2005

효과적인 교육을 위한 다양한 연구들이 진행되어 왔지만, 이러한 결과를 실제 수업현장에 적용하기에는 많은 시간과 노력 등이 필요하여 잘 활용되지 못하고 있는 것이 현실이다. 따라서 교실 수업, 특히 지필 위주의 수업에서는 여전히 따분해하며 집중하지 못하는 학생들을 흔하게 볼 수 있다. 본 연구에서는 학생들의 적극적 참여를 통한 효과적인 교육을 구현하기 위한 방안으로 구성주의적 교수학습 방안과 정보 기술을 기반으로 교수자와 개별 학생들 사이의 실시간 상호작용을 극대화시키는 교수방안을 제안하고자 한다. 구현 사례로 통계적 추정의 표본추출 단원에 적용할 수 있는 교수학습 모형과 시스템을 제시한다.

• Proceedings of the KSRS Conference
• /
• 2006.03a
• /
• pp.359-364
• /
• 2006

최근 GIS를 비롯한 공간정보기술의 눈부신 발전으로 국토전반에 대한 정보화와 과학적인 관리기법이 확산되고 있다. 이에 효율적인 산림관리를 위하여 위성영상과 지리정보시스템을 이용한 산림정보가 구축되고 있으나, 산림지역이라는 지리특성상 현장 출입이 제한적이고 위험하여 정보수집에 어려움이 있다. 본 연구에서는 효율적인 산림정보구축을 위하여 남제주군을 대상으로 고해상도 위성영상과 공간정보추출언어인 SML(Spatial Modeler Language)을 이용하여 다양한 모델별 산림지역의 임상정보를 추출, 임상분류를 위한 모델링 기법을 개발하였다. 또한 l/5,000도엽을 기준으로 표본점을 배치하고 임상, 하층식생, 입지 및 토양에 대한 현지조사를 수행하였으며, 추출자료와의 상관성 분석 및 검증을 통하여 공간정보 기술을 이용한 산림정보 구축의 기반을 마련하고자 하였다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2004.05b
• /
• pp.272-276
• /
• 2004

이수기와 같이 장기적인 관점에서 저수지운영을 해야 하는 관리자는 해당 기간동안의 이익을 최대화하는 전략을 필요로 한다. 이를 위해서는 미래 유입량의 불확실성을 고려한 최적화 모형에 근거한 운영률을 수립해야 할 것이다. 본 연구에서는 금강수계의 이수기를 대상으로 추계학적 최적화 기법인 표본 추계학적 동적계획법(Sampling Stochastic Dynamic Programming)을 적용하여 최적 연계운영 시스템을 개발하였다. 본 연구를 통해 개발된 모형은 상용프로그램인 CSUDP와의 비교를 통해 검증되었으며 이를 기반으로 과거자료를 이용한 SSDP/Hist모형과 앙상블 유량예측(Ensemble Streamflow Prediction)을 이용한 SSDP/ESP모형을 개발하여 두 모형의 장${\cdot}$단점을 비교 분석하였다. 발전부분은 두 모형이 비슷하였으나 용수공급 측면에서는 SSDP/ESP가 SSDP/Hist 보다 우수함을 알 수 있었다.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.17 no.4
• /
• pp.69-85
• /
• 2014

Carbon stocks of NFI plots can be accurately estimated using field survey information. However, an accurate estimation of carbon stocks in other unsurveyed sites is very difficult. In order to fill this gap, various spatial information can be used as an ancillary data. In South Korea, there is the 1:5,000 forest type map that was produced by digital air-photo interpretation and field survey. Because this map contains very detailed forest information, it can be used as the high-quality spatial data for estimating carbon stocks. In this study, we compared three upscaling methods based on the 1:5,000 forest type map and 5th national forest inventory data. Map algebra(method 1), RK(Regression Kriging)(method 2), and GWR(Geographically Weighted Regression)(method 3) were applied to estimate forest carbon stock in Chungcheong-nam Do and Daejeon metropolitan city. The range of carbon stocks from method 2(1.39~138.80 tonC/ha) and method 3(1.28~149.98 tonC/ha) were more similar to that of previous method(1.56~156.40 tonC/ha) than that of method 1(0.00~93.37 tonC/ha). This result shows that RK and GWR considering spatial autocorrelation can show spatial heterogeneity of carbon stocks. We carried out paired t-test for carbon stock data using 186 sample points to assess estimation accuracy. As a result, the average carbon stocks of method 2 and field survey method were not significantly different at p=0.05 using paired t-test. And the result of method 2 showed the lowest RMSE. Therefore regression kriging method is useful to consider spatial variations of carbon stocks distribution in rugged terrain and complex forest stand.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.19 no.4
• /
• pp.169-185
• /
• 2016

Recently mean stand height has been added as new attribute to forest type maps, but it is often too costly and time consuming to manually measure 9,100,000 points from countrywide stereo aerial photos. In addition, tree heights are frequently measured around tombs and forest edges, which are poor representations of the interior tree stand. This work proposes an estimation of mean stand height using an image-based point cloud, which was extracted from stereo aerial photo with FUSION S/W. Then, a digital terrain model was created by filtering the DSM point cloud and subtracting the DTM from DSM, resulting in nDSM, which represents object heights (buildings, trees, etc.). The RMSE was calculated to compare differences in tree heights between those observed and extracted from the nDSM. The resulting RMSE of average total plot height was 0.96 m. Individual tree heights of the whole study site area were extracted using the USDA Forest Service's FUSION S/W. Finally, mean stand height was produced by averaging individual tree heights in a stand polygon of the forest type map. In order to automate the mean stand height extraction using photogrammetric methods, a module was developed as an ArcGIS add-in toolbox.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.13 no.6
• /
• pp.25-33
• /
• 2014

This study investigates the reliability of DSRC-based traffic information system on the typical urban arterial with the minimum sample size method. VISSIM has been employed to calculate the required sample size. After comparing the number of hi-pass vehicles recorded from DSRC and the required sample size, this study found that the interrupted traffic state tends to generate more outliers than the uninterrupted one, the lack of the number of vehicles completely passing links with multiple driveways makes it difficult to estimate the reliable traffic information, the traffic information during peak hour is relatively more reliable than that during off-peak hour, and the reliability of DSRC-based traffic information system depends on the significance level in calculating the sample size. The driveway density and traffic signal operation due to the individual link length significantly affects the required sample size, resulting in determining the reliability of the DSRC-based traffic information system.