• The Korean Journal of Applied Statistics
• /
• v.35 no.2
• /
• pp.177-188
• /
• 2022

The support vector machine (SVM) has been successfully applied to various classification areas with a high level of classification accuracy. However, it is infeasible to use the SVM in analyzing massive data because of its significant computational problems. When analyzing imbalanced data with different class sizes, furthermore, the classification accuracy of SVM in minority class may drop significantly because its classifier could be biased toward the majority class. To overcome such a problem, we propose the DOC-SVM method, which uses divide-oversampling and conquers techniques. The proposed DOC-SVM divides the majority class into a few subsets and applies an oversampling technique to the minority class in order to produce the balanced subsets. And then the DOC-SVM obtains the final classifier by aggregating all SVM classifiers obtained from the balanced subsets. Simulation studies are presented to demonstrate the satisfactory performance of the proposed method.

• Journal of Sasang Constitutional Medicine
• /
• v.13 no.3
• /
• pp.15-22
• /
• 2001

Revised QSCCII is made depend on revised soeum scale, revised taeum scale and revised soyang scale, and used in KyungHee University Medical Center. The purpose of this study is to examine validity of revised QSCCII and to analyse the tendency of respondents of revised QSCCII. For this study, the objects are 57 patients who response to QSCCII and revised QSCCII in Oriental Medicine & Western Medicine Cooperative Health Examination Center KyungHee University Medical Center. a systematic analysis has been made. As a results, we have found that revised QSCCII have a validity. there is agreement of Tendency of the respondent of QSCCII with revised QSCCII. we hope that suitable questionnaire will be developed for the research in the future.

• The Korean Journal of Applied Statistics
• /
• v.27 no.3
• /
• pp.357-371
• /
• 2014

There are many studies related to imbalanced data in which the class distribution is highly skewed. To address the problem of imbalanced data, previous studies deal with resampling techniques which correct the skewness of the class distribution in each sampled subset by using under-sampling, over-sampling or hybrid-sampling such as SMOTE. Ensemble methods have also alleviated the problem of class imbalanced data. In this paper, we compare around a dozen algorithms that combine the ensemble methods and resampling techniques based on simulated data sets generated by the Backbone model, which can handle the imbalance rate. The results on various real imbalanced data sets are also presented to compare the effectiveness of algorithms. As a result, we highly recommend the resampling technique combining ensemble methods for imbalanced data in which the proportion of the minority class is less than 10%. We also find that each ensemble method has a well-matched sampling technique. The algorithms which combine bagging or random forest ensembles with random undersampling tend to perform well; however, the boosting ensemble appears to perform better with over-sampling. All ensemble methods combined with SMOTE outperform in most situations.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.10a
• /
• pp.314-316
• /
• 2009

본 논문에서는 사상체질분류검사 설문지를 이용하여 사상체질을 진단할 때, 진단의 정확도를 향상시키기 위한 사상체질 분류 함수를 개발하기 위하여 퍼지 분류기를 이용한다. 본 연구에서 사용하는 데이터는 9개 한의과대학의 10개 부속한방병원에서 치료를 받은 환자들 중 각 병원의 사상체질전문의로부터 체질진단을 받고 최소한 4주 이상 사상체질 처방을 사용한 후 주 증상이 전반적으로 호전되어 체질이 확인된 환자 1,914명을 대상으로 하고 있다. 본 연구는 사상체질의학의 광제설을 토대로 환자의 성별을 분리 하였을 뿐만 아니라, 비만도를 추가적으로 분류하였으며, 체형기상, 용모사기, 성질재간, 병증약리 중 체형기상을 토대로 분류하였으며, 사상체질을 판별할 수 있도록 설계되고 구현되었다.

• Journal of Convergence for Information Technology
• /
• v.10 no.4
• /
• pp.98-103
• /
• 2020

The purpose of this study was to determine whether the results of KTAS(Korean Triage and Acuity Scale) triage classifier differ according to the occupations. We analyzed a total of 10,960,359 cases of data sent to the NEDIS from January 1st, 2016 to December 31th, 2017. The triage classifier were MD(Medical Doctor), R(Resident), INT(Intern), GP(General Practitioner), RN(Registered Nurses) and EMT(Emergency Medical Technician). The consistency between the initial triage and final triage results was the highest GP(98.9%) and the lowest INT(80.2%). The results of over-triage classification was the lowest by GP(0.6%) and the highest for INT(16.0%). Also, the results of under-triage classification was the lowest by MD, EMT(0.4%) and the highest for INT(3.8%). The results of KTAS triage classifier significantly differ from according to the occupations(p<0.001). Triage classification should not differ from according to occupations and skill. It is necessary to strengthen the classifier's capacity for accurate triage classifications.

• Journal of The Geomorphological Association of Korea
• /
• v.19 no.3
• /
• pp.1-22
• /
• 2012

The classification scheme of estuaries can be divided into two categories: qualitative classification based on geomorphic characteristics and quantitative classification based upon the physical properties of water body. While simple and intuitive scheme of the former is difficult to quantify, the latter is not easy to apply due to the lack of data. A classification scheme based on morphological convergence is very promising because it only requires easily accessible data such as width and depth of channels, as well as it can characterize estuaries in terms of tidal propagation. Thus, this paper examines the classification scheme based on estuarine morphological convergence using depth and width data obtained from 19 major Korean estuaries. Morphological convergence for each estuary was estimated with the estuarine length, width and depth data to get the convergence parameters, which includes the degree of funneling ${\nu}$ and the dimensionless estuarine length $y_0$. The transfer function ${\xi}({\nu},ky)$ is then deduced analytically from 1D depth-integrated hydrodynamic momentum equation and continuity equation for estuarine shapes. Tidal response of each estuary is finally calculated using ${\nu}$, $y_0$ and ${\xi}({\nu},ky)$ for comparison and classification. The 19 Korean estuaries were classified into three groups: tidal amplitude-dominated estuaries with standing wave-like tidal response (group 1), current-dominated estuaries with progressive wave-like tidal response (group 2), and the intermediate group (group 3) between groups 1 and 2. The sensitivity analysis revealed that uncertainties in determining the estuarine length can have a critical effect upon the results of classification, which indicates that the reasonable determination of the estuarine length is of critical importance. Once the estuarine length is feasibly determined, depth-convergence can be neglected without any negative effect on the classification scheme, which has an important ramification on the wide applicability of the classification scheme.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.10a
• /
• pp.22-24
• /
• 2003

작은 데이터에서 베이지안망 분류기(Bayesian network classifier, BNC)를 학습할 때, 과대적합(overfitting)으로 인한 일반화 성능의 저하가 초래된다 이런 경우, 베이지안 모델 평균화(Bayesian model averaging, BMA)는 모델 자체에 대한 불확실성을 분석 과정에서 고려함으로써, 성능 저하를 피할 수 있는 수단을 제공한다. 본 논문에서는 BNC의 BMA의 작은 데이터에 대한 성능을 평가 및 분석한다. 특히, 노드의 순서에 대한 평균화의 효과가 연구된다. 인공데이터에 대한 실험 결과, 노드의 순서가 BNC의 BMA의 분류 성능에 미치는 영향은 지대하며, 이는 데이터의 크기가 극히 작은 경우의 성능 저하에 직접적인 원인이 된다.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.6
• /
• pp.1-7
• /
• 2022

The study was conducted to determine the effectiveness of the triage training program utilizing pre-and post-training experiments designed for 119 emergency medical services teams. Objectives: This study evaluated the effectiveness of triage training programs on the accuracy of triage performed by 119 emergency medical services team staff who participated in the triage training program. Behavior: Participants in this study included 119 of the 166 EMS staff. In this program, a modified START triage consisting of a 20-minute theoretical presentation was presented to the participants. Data were analyzed using SPSS 21.0. Results: A significant increase in triage accuracy for 119 EMS teams(p<.001). And undertriage showed a significant decrease(p<.001). In addition, overtriage showed a decrease but was not statistically significant. Conclusions: The results obtained from this study showed that the triage training program was effective in improving the accuracy of the triage of multiple injury patients or disaster victims when presented to the 119 emergency medical services team. Therefore, these results suggest that it would be helpful to add triage training to the fire department's formal training program.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2009.05a
• /
• pp.1265-1271
• /
• 2009

본 연구에서는 강우의 시 공간적 분포의 불규칙한 변동성을 고려한 강우-유출예측모형을 위해 인공신경망(Artificial Neural Networks: ANNs)의 기법의 일종인 자기조직화(Self Organizing Map: SOM) 이론과 역전파 학습 알고리즘(Back Propagation Algorithm: BPA) 이론을 복합적으로 이용하였다. 기존의 인공신경망 연구에서 야기된 저 갈수기의 유출량에 대한 과대평가, 홍수기의 유출량에 대한 과소평가, 예측값이 연속적으로 선행 유출량을 나타내는 Persistence 현상을 해결하기 위하여 패턴분류 성능을 지닌 SOM 이론을 예측모형의 전처리 과정으로 이용하였다. 먼저, 본 연구에서 제안한 방법은 SOM에 의해 강우-유출 관계를 분류하고, SOM에 의한 분류에 따라 각각의 모형을 구성한다. 개별적으로 구축된 모형은 유출량의 예측을 위해 각각의 양상에 따라 분류된 자료를 이용한다. 결과적으로 본 연구에서 제안한 방법은 과거의 인공신경망의 일반적인 적용에 의한 결과보다 더 나은 예측능력을 보여주었으며, 더불어 유출량의 과소 및 과대추정과 Persistence 현상과 같은 문제점이 나타나지 않았다. 또한 강우량 및 유출량의 범위에 제한을 받지 않는 강우-유출예측 모형의 개발 및 홍수기로부터 갈수기까지의 보다 넓은 범위의 유출량의 예측에 기여할 것으로 기대된다.

• Proceedings of the KOR-KST Conference
• /
• 1998.10a
• /
• pp.31-40
• /
• 1998

본 연구의 목적은 기존의 Cell Transmission(1994, Daganzo) 교통류 이론을 기반으로 동적통행배정 모형을 개발하는 것이다. 이 모형은 동적 O-D 발생모듈, HOV 차선모듈, 분류부 분할모델, 링크비용함수 모듈, 최단경로 탐색 모듈등으로 구성된다. 이 모델에서 적용하는 교통류 모델은 각 링크를 동일한 특성을 가지는 셀로 구분하여 셀내의 진입시간과 진출시간을 계산하여 링크비용을 계산하는데 이것은 비용의 과대·과소 추정을 피할 수 있으며 교통지체 현상을 현실적으로 표현해 줄 수 있는 장점이 있다. 또한 HOV 차선 모듈에 의해 수단별 교통류 진행 및 비용고려가 가능하며 HOV 차선의 평가 및 분석이 가능하다. 기존의 동적통행배정모형은 매 시간대별 출발지에서 균형상태를 추구하는 통행배정기법을 사용하고 있지만 이 모델은 분류되는 노드를 가상의 출발점이라고 가정하여 각 시간대별로 최단경로를 탐색하여 균형상태를 추구해나가는 기법을 적용하고 있다. 각 셀별 차량을 목적지별, 차종별, 대기시간별로 추적하여 진행시키며 분류부에서는 최단경로를 탐색하여 배분된다. 또한 진행하고자 하는 셀의 용량과 현재 셀의 밀도를 고려함으로서 용량제약 하에서의 동적통행배정모형을 적용하고 있다. 이 모형은 고속로의 합류부 및 분류부의 교통특성을 세밀히 분석할 수 있으며, TCS 및 램프미터링과 접목하여 고속도로 운영에 이용될 수 있으며, 고속도로 중·장기적인 계획에 이용될 수 있다.