http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
강병갑,허태영 대한교통학회 2018 대한교통학회 학술대회지 Vol.78 No.-
교통사고는 실제 도로 위에서 대부분 발생한다. 교통사고와 같은 네트워크 구조를 기반으로 하는 네트워크 공간 데이터 분석을 계산상의 편의성과 네트워크 거리의 근사성의 이유로 평면 기반의 유클리드 거리 방법을 사용하는 경우가 많으며, 이 방법을 이용하여 추정하면 군집성을 과대 추정하여 잘못된 결과를 도출할 수 있다. 본 연구에서는 공간 점 패턴 분석으로 평면과 네트워크의 K, G, F함수를 이용하여 요일 및 교통사고 유형별로 군집성 결과를 비교하였으며, AIC값으로 모형의 적합성을 평가하였다. 국내 서울 강남구와 국외 뉴욕의 교통사고에 대한 평면과 네트워크 공간 점 패턴 분석을 비교한 결과, 평면에서 군집성을 과대 추정하는 것을 확인하였으며, 평면과 네트워크에서 군집성 결과가 다양하게 나타나는 것을 확인하였다. 또한, 네트워크 모형의 AIC값이 평면보다 큰 차이를 보여 네트워크 기반 공간 점 패턴 분석 모형이 더 적합한 것을 확인하였다. 본 연구에서 제안한 네트워크 기반공간 점 패턴 분석을 교통사고에 적용하여 발생위치에 대한 공간 군집성 결과를 도출하여 교통사고 발생 지점을 정확하게 파악하고 문제점을 해결하는데 활용할 수 있을 것이다.
네트워크 기반 공간 점 패턴 분석을 이용한 교통사고 발생 군집성 분석
강병갑,허태영 한국자료분석학회 2017 Journal of the Korean Data Analysis Society Vol.19 No.5
The purpose of this study is to compare planar and network spatial analysis of traffic accidents for Gangnam-gu, Seoul. Total 41,662 traffic accident data collected from all over Seoul metropolitan area in 2010. This data includes information on a total 31 variables such as the location of the accident, the type of car, accident and road, weather condition, number of dead and injured, sex and age. Spatial clustering is analyzed by spatial point pattern analysis using K, G and F function based on linear network for 3,536 traffic accidents in Gangnam-gu. The results of the analysis of planar and network spatial pattern in K, G and F function are the same, but the difference of AIC value is smaller in network spatial pattern analysis. Our results are expected that the use of the network spatial point pattern analysis results can be used as a more accurate and correct information providing method to identify the trend of traffic accidents and solve the problems of the location of traffic accidents in the future. 본 연구에서는 교통사고 발생위치에 대해 기존의 평면 기반이 아닌 네트워크 기반 공간 점 패턴 분석을 통해 보다 정확한 교통사고 위험도 평가를 목적으로 하고 있다. 서울특별시 교통사고 발생 건수 총 41,662건 중 교통사고 발생이 빈번한 강남구에서 발생한 교통사고 3,536건에 대해 네트워크 기반 K, F, G 함수를 이용하여 공간 점 패턴 분석을 통해 공간적 군집성을 파악하였으며 기존의 평면공간분석 결과와 비교하였다. 분석을 위한 자료는 사고지점을 나타내는 경/위도, 자동차 유형, 성별, 나이, 사고유형, 도로종류, 기상상태, 사망자수, 중상자수 등 총 31개의 변수에 대한 정보가 포함되어 있다. 결과적으로 K, G, F 함수별 모두 군집성 판별에서는 평면과 네트워크 공간 패턴 분석에서 동일한 결과를 제공하지만, 네트워크 기반 공간 패턴 분석의 정확도가 높은 것으로 확인되었다. 본 연구에서 제안한 네트워크 기반 분석을 통해 교통사고 발생위치에 대한 공간 군집성 분석 결과를 이용하여 교통사고 발생 경향을 파악하고 문제점을 해결하는데 활용할 수 있을 것으로 판단된다.
강병갑,고미미,정지연,이주아,Kang, Byoung-Kab,Ko, Mi Mi,Jung, Jeeyoun,Lee, Ju Ah 대한예방한의학회 2017 대한예방한의학회지 Vol.21 No.1
Objectives : The aims of the current study were to assess the reliability and validity of the CoRe-Ditec-BS Questionnaire-II (BSQ-II; Blood stasis questionnaire-II) with 8 items including 5 items related women added to BSQ-I that was consisted of 36 items and developed in 2013. Methods : Between May 2014 and November 2014, 411 patients from 3 traditional Korean medical hospitals were asked to complete the BSQ-II. Each patient was independently diagnosed with BSS by two traditional Korean medical physicians from the same site. We estimated the internal consistency using Cronbach's ${\alpha}$ coefficient, the discriminant validity using the means score of BSS, and the predictive validity using logistic regression (sensitivity and specificity). Results : The BSQ-II had satisfactory internal consistency (Cronbach's ${\alpha}$ coefficient=0.765) and validity, with significant differences in the mean scores between the BSS($63.60{\pm}9.56$) and non-BSS groups($48.36{\pm}5.93$). The area under the receiver operating curve was about 98%, and the sensitivity and specificity were 91.4% and 94.9%, respectively. Conclusions : These results suggest that the CoRE-Ditec-BSQ-II is more reliable and valid instrument for estimating BSS than BSQ-I.
강병갑(Byeongkab Kang),고호연(Hoyeon Go),강경원(Kyungwon Kang),박세욱(Saewook Park),김정철(Jeongchul Kim),고미미(Mimi Go),김보영(Boyoung Kim),설인찬(Inchan Seol),이인(In Lee),조현경(Hyunkyung Jo),최선미(Sunmi Choi) 한국한의학연구원 2007 한국한의학연구원논문집 Vol.13 No.2
Background and Purpose : The purpose of this study was to investigate that which symptoms are adequate indicator of the Dampness-Phlegm pattern in the stroke patients. Methods : In the time period Jul. 2005 to Sep. 2006, 136 patients with a first-ever stroke admitted in the department of Internal Medicine of Daejeon University Oriental Medical Hospital in Daejeon city, Wonkwang Oriental Medical Hospital in Iksan, JeonJu city were included. Patients were hospitalized within 3 months after the onset of stroke. Stroke patients had been interviewed by resident who studied standard operation procedures in Fundamental Study for Standardization and Objectification of Differentiation and Pattern Identification of Syndrome of Oriental Medicine for Stroke. Gi-deficiency patients was confirmed by medical specialist diagnosis, resident diagnosis, case report form analysis without a dissenting voice. Results : Dampness-Phlegm group included 37 case, Non Dampness-Phlegm group 45 case out of 136 patients. white tongue coating, slippery purse, yellowish complexion, enlarged tongue, swollen tongue were higher among Dampness-Phlegm group. Dampness-Phlegm and Non Dampness-Phlegm patients do not significantly differ in heavy sensation in the head, voice with sputum, teeth printed tongue, borborygmus, dizziness with nausea. Conclusions : This study was insufficiency because sample size is very small. More data from prospective cohort studies will help to Korean Standard Differentiation of the Symptoms and Signs for the Stroke.
강병갑(Byeongkab Kang),강경원(Kyungwon Kang),박세욱(Saewook Park),김보영(Boyoung Kim),김정철(Jeongchul Kim),고미미(Mimi go),설인찬(Inchan Seol),조현경(Hyunkyung Jo),이인(In Lee),최선미(Sunmi Choi) 한국한의학연구원 2007 한국한의학연구원논문집 Vol.13 No.2
The purpose of this study was to diagnosis that what patterns identification using the statistical method. Discriminant analysis using the medical specialist and resident pattern identification agree case in stroke patients within 1 month of onset. The agreement rate of dificiency of Gi(75%), heat-transformation(74%), dampphlegm syndrome(69%), deficiency of Eum(51 %) and syndrome of blood stagnation(43%) are respectively 0.75, 0.74, 0.69, 0.51 and 0.43 in medical specialist and using linear discriminant function pattern identification are same. The study of inspection, pulse feeling and palpitation will be continued to evaluate concordance rate. Discrimination model will be make to get higher Accuracy and prediction, it means becomes the help in pattern identification diagnosis objectivity and scientific.
강병갑(Kang Byeongkab),문진석(Moon Jinseok),최선미(Choi Sunmi) 한국한의학연구원 2007 한국한의학연구원논문집 Vol.13 No.1
The high position condition 10 escape which the obesity person appeals. Obesity Pattern-Identification question it will yell and 243 subjects which to the obesity in the patient of 517 subjects which draw up correspond. (longitude obesity 153 person, altitude obesity 90 person) against it analyzes. In order to analyze the reliability of the items which diagnose each Pattern-Identification it used Cronbach alpha coefficient and escape it did the alpha of each item. Alpha value of each Pattern-Identification than appears more highly the item which it will be able to consider an elimination in the item which decreases a reliability. In that phlegm-retention syndrome is bigger alpha coefficient 0.784 than 'meal quantity is few'(0.787) a possibility of seeing in the item which decreases a reliability to the place where it diagnoses it puts in phlegm-retention syndrome.