• 산경연구논집
• /
• 제11권7호
• /
• pp.29-40
• /
• 2020

Purpose: In modern society, many urban problems are occurring, such as aging, hollowing out old city centers and polarization within cities. In this study, we intend to apply big data and machine learning methodologies to predict depression symptoms in the elderly population early on, thus contributing to solving the problem of elderly depression. Research design, data and methodology: Machine learning techniques used random forest and analyzed the correlation between CES-D10 and other variables, which are widely used worldwide, to estimate important variables. Dependent variables were set up as two variables that distinguish normal/depression from moderate/severe depression, and a total of 106 independent variables were included, including subjective health conditions, cognitive abilities, and daily life quality surveys, as well as the objective characteristics of the elderly as well as the subjective health, health, employment, household background, income, consumption, assets, subjective expectations, and quality of life surveys. Results: Studies have shown that satisfaction with residential areas and quality of life and cognitive ability scores have important effects in classifying elderly depression, satisfaction with living quality and economic conditions, and number of outpatient care in living areas and clinics have been important variables. In addition, the results of a random forest performance evaluation, the accuracy of classification model that classify whether elderly depression or not was 86.3%, the sensitivity 79.5%, and the specificity 93.3%. And the accuracy of classification model the degree of elderly depression was 86.1%, sensitivity 93.9% and specificity 74.7%. Conclusions: In this study, the important variables of the estimated predictive model were identified using the random forest technique and the study was conducted with a focus on the predictive performance itself. Although there are limitations in research, such as the lack of clear criteria for the classification of depression levels and the failure to reflect variables other than KLoSA data, it is expected that if additional variables are secured in the future and high-performance predictive models are estimated and utilized through various machine learning techniques, it will be able to consider ways to improve the quality of life of senior citizens through early detection of depression and thus help them make public policy decisions.

• 정보처리학회논문지:컴퓨터 및 통신 시스템
• /
• 제6권9호
• /
• pp.377-384
• /
• 2017

발열 때문에 더이상 회로 집적도를 높일 수 없기 때문에 단일 코어 프로세서의 성능 향상은 한계에 달했다. 그래서 코어를 여러 개 사용하는 멀티 코어, 매니 코어 형태의 프로세서가 등장했으며 병렬 프로그래밍이 중요해졌다. 이러한 상황에서 병렬 프로그래밍에 여러 장점이 있는 순수 함수형 언어 Haskell이 주목받고 있다. Haskell은 식 계산 방식에서 이미 병렬성이 내재되어 있으며 병렬 구조를 지원하는 모나드 도구를 제공한다. 그런데 Haskell 병렬 프로그램의 성능은 메모리 재사용 시스템을 포함한 실행시간 시스템에 큰 영향을 받는다. 이미 Haskell이 제공하는 메모리 프로파일링 도구로 GC-tune이 있지만, GC-tune은 가능한 모든 GC 옵션에 대해 프로그램 실행 시간을 반복 측정하기 때문에 GC 조정 시간이 너무 오래 걸린다. 그래서 본 연구에서는 기본적인 분할 정복법을 이용해서 GC-tune의 탐색 영역을 매 단계마다 4분의 1로 줄이는 방법을 제안한다. 제안하는 방법을 두 가지 병렬 프로그램(극대 독립 집합 프로그램과 K-평균 프로그램)에 적용한 결과, 평균 98%의 정확도로 실행 시간을 평균 7.78배 단축시켰다.

• 대한인간공학회지
• /
• 제34권3호
• /
• pp.247-264
• /
• 2015

Objective: The purpose of this study is to determine the priority of information presentation and the effective menu type to be placed in the center of a gauge cluster display for commercial trucks and to present a set of ergonomic designs for the gauge cluster display. Background: An effective ergonomic design is specifically needed for the development of the gauge cluster display for the commercial trucks, because more diverse and heavier information is delivered to truck drivers, compared to the information to passenger car drivers. Method: First, all the information that must be shown on the commercial truck display was collected. Then, the severity, frequency of use, and display design parameters were evaluated for those information by commercial truck drivers. Next, an analysis on the information attributes and the heuristic evaluation utilizing the display design principles were carried out. According to the results, a design alternative of the main screen to be displayed was constructed by priority. A comparative analysis between the alternative and existing main screens was also conducted to see the efficacy of the designs. Lastly, we conducted an experiment for the selection of menu type. The experiment was conducted using the driving simulator with an eye-tracking device. The independent variables were four types of the menu reflecting the commercial truck characteristics such as grid type, icon type, list type, and flow type. We measured preference, total execution time, the total duration of fixation on the gauge cluster area, and the total number of fixation on the gauge cluster area as dependent variables. Results: Four types of driver convenience information and six types of driver assistance information were selected as the information to be placed primarily on the main screen of the gauge cluster. The Grid type was the most effective among the menu types. Conclusion: In this study, the information that appears on the main screen of the display, the division of the display and the design of the menu type for commercial truck drivers were suggested. Application: This study is expected to be utilized as guidelines on the ergonomic design of a gauge cluster display for commercial trucks.

• 대한간호학회지
• /
• 제21권1호
• /
• pp.89-116
• /
• 1991

The purpose of this study was to describe mother -infant interacting behavior patterns related to newborn infant feeding and to explore the mother's cultural belief about their infant. The data collection was conducted by observation and interview. Twenty-five mothers and their newborn infants who were normally delivered and were also planned to breast feed were comprised as the subjects of this study. All subjects were interviewed and observed individuaily at 1 to 5 days after the delivery at the hospital, mid -wife's clinic, Maternal Child Health Center and their home throughout the country from remote area to big city, The observation data were recorded with symbolic letter on a recording sheet newly developed as a result of preliminary study. The interview data were taperecorded and then recorded in narrative form. Mother - infant interaction behaviors in early feeding period were analyzed based on 19 analytic sub-categories and their composing elements. Unit of analysis were mother, infant and mother -infant dyad. 8 analytic categories draw from the data. Each were preparation, instrument, interaction inducing, evaluation referred to mother's behavior, preparation, instrument, interaction inducing referred to infant's behavior and synchronic behaviors referred to mother - infant dyad. Frequencies of behavior items based on the categories were converted to percent. The result showed that in mother's preparation behavior, the breast condition of Korean mother can be an affecting factor for mother - infant interaction during feeding, and vocalization behavior was observed most frequently in interaction inducing behavior while the least frequent behavior observed was contacting. Subcultural characteristics of mother - infant interaction behaviors were analyzed for their relationships between groups of mothers who have lived in remote area vs urban area, and who were multipara vs primipara. Using a chi -square test, there were statistically significant relationships in the activity of psychological readiness in preparation behavior and the movement of extremities for the position of instrumental behavior in both groups. However, interaction inducing behaviors were not related with statistical significance in any set of groups. Accomplishment of marriage, bonding and emotional mediation of family members were the categories related to mother's cultural belief about the infant in aspect of functional values. Infant at birth is considered little more than a biological organism without social capabilities. Although the newborn infant is still be attached to his mother, he makes his mother extend her territoriality. The mother's interacting behavior toward her infant based on those beliefs appeared task oriented, separative behavioral series. On the other hand, it was seen that infant reacted independently to his mother's behavior by the in-nate perceptual abilities. Those independent behavioral series of mother and infant on the feeding situation were synchronized at any moment. Nurses are In a unique position to teach mothers about their infant's capabilities and help reducing some of uncertainty about infant's behaviors. Study results indicated that the informations infant's social capabilities and breast feeding should be given to the mothers. The results of this study have several implications for nursing. First, the study results will be used as fundamental resources for the development of the assessment tool about the early mother - infant interaction. Second, the results could be a relevant information in the fied. I of maternal child nursing education as real and useful data. Third, the behavioral patterns of early mother - infant interaction which were classified based on the qualitative analysis could be used for nursing theory development as very fundamental data.

• 전자공학회논문지
• /
• 제50권7호
• /
• pp.76-84
• /
• 2013

순시(instantaneous) TDOA (time difference of arrival)와 FDOA (frequency difference of arrival)를 이용한 위치추정 방법은 추가적인 측정값 획득을 통해 정확도 향상을 도모할 수 있으며, 이를 위해서는 동시에 운용되는 수신단의 수를 증가하여야 한다. 하지만 전자전 환경에서 수신단 수의 증가는 아군의 피탐확률(probability of intercept) 상승으로 인한 전력 손실을 야기할 수 있고, 수신단 간의 데이터 링크 및 시각동기화와 같은 과정에 대한 추가적인 고려가 필요하다. 따라서 본 논문에서는 이격된 2개의 이동 수신단만을 운용하여 연속적으로 다수의 TDOA와 FDOA 정보를 측정하고, 이를 이용하여 고정 신호원의 위치를 추정하는 방법을 제안한다. 이 경우 매 측정 순간마다 독립된 수신단 쌍(pair)이 추가되므로 각 수신단 조합은 서로 다른 기준 수신단을 가지게 된다. 그러므로 모든 수신단 쌍이 동일한 기준 수신단을 공유해야하는 QCLS (quadratic correction least squares) 방법을 적용할 수 없다. 이러한 이유로 본 논문에서는 비선형 LS 최적해를 반복계산을 통해 얻어내는 Gauss-Newton 기법을 적용한다. 또한 모의실험을 통해 획득된 TDOA와 FDOA의 수가 증가함에 따른 위치추정 결과의 RMSE (root mean square error)값과 CRLB (Cramer-Rao lower bound)를 비교하고, CEP (circular error probable) 평면을 도시하여 2차원 공간상에서의 기대 추정 성능을 분석한다.

• 항공우주정책ㆍ법학회지
• /
• 제22권1호
• /
• pp.75-103
• /
• 2007

본 논문은 동북아시아에서의 우주협력에 장애가 되는 것을 극복하고자 정치적 상황을 검토함과 아울러 규범적인 체계의 수립에 대해서 검토하고자 한다. 그러한 협력의 장애가 되는 것중의 하나는 MTCR을 중심으로 하여 미국의 영향력 이라고 할 수 있다. 한국과 중국의 발사계약이 성사되지 않은 것도 그로 인한 것이었다. 중국이 MTCR 체제와 융합하는 것이 동북아시아에서의 우주협력을 활성화하기 위한 초석이 될 수도 있다. Helsinki 사례가 인권, 평화 및 환경 문제를 다루는 포괄적인 규범 체계를 제공하였듯이, 한반도의 안정과 평화를 둘러싸고 수립될 수 있는 동북아시아에서의 평화 체제가 우주부문에서의 협력과 같은 광범위한 문제를 다룰 수도 있을 것이다. 한국이 동북아시아지역에서 제3의 우주개발국가로서 성장하면 중국과 다른 열강과의 관계에 있어서 균형자로서의 역할을 할 수도 있을 것이다. 한국은 우주개발을 통해서 획득한 자료등을 아무런 정치적 이해 관계없이 주변국들과 공유하면서 우주협력을 이끌어낼 수도 있다. 일본이 주도하는 APRSAT가 재난방지위성시스템으로서 Sentinel을 제공하면서 주변 국가들의 협력을 이끌어 낸 것이 하나의 예가 될 수도 있다.

• 한국식품영양과학회지
• /
• 제41권5호
• /
• pp.695-699
• /
• 2012

본 연구에서는 양파가공부산물인 양파껍질로부터 반응표면분석법을 이용하여 유효성분인 flavonoid 물질의 최적 추출조건을 설정하고자 하였다. 양파껍질의 일반성분은 수분 4.68%, 회분 6.67%, 조단백질 및 조지방이 5.03%, 0.71%이었고, 무기질은 Ca이 20.07 g/kg으로 가장 많은 함량을 차지하였고 K이 5.54 g/kg, Mg 2.17 g/kg의 순으로 함량이 많았다. 중심합성계획법에 의한 flavonoid 물질의 최적 추출조건은 다중회귀분석 및 능선분석 결과, 결정계수는 0.772이었으며, 적합결여검증(p>0.05)에서는 0.278, 전체적인 모형에서 유의성(p<0.05)을 보여 반응모형이 적합한 것으로 판단되었다. 따라서 최대점을 만족하는 각 변수의 조건은 추출용매의 농도는 70%, 추출온도 $40^{\circ}C$, 추출용매의 pH 5.3, 원료에 대한 추출용매의 비율은 1:63(w/v)이었다.

• 한국산림과학회지
• /
• 제102권3호
• /
• pp.437-445
• /
• 2013

본 연구는 환경인자를 이용하여 우리나라에 생태권역별로 분포하는 상수리나무의 지위지수 추정식을 개발하고 적지면적 및 분포를 추정하기 위해 수행하였다. 이를 위해 산림입지도와 전자기후도로부터 산림생산력에 영향을 미칠 것으로 판단되는 19개의 기후변수를 포함한 총 48개 환경인자를 도출한 후, 최적 조합에 의해 지위지수 추정식을 개발하였다. 최종 생태권역별 상수리나무의 지위지수 추정식에는 각각 4~6개의 환경인자가 독립변수로 사용 되었고, 지위지수 추정식의 설명력을 나타내는 결정계수는 0.30~0.41의 범위에 있는 것으로 분석되었다. 또한 모형의 평균편의, 정도(精度), 표준오차의 3가지 평가통계량을 이용하여 검증을 실시한 결과 비교적 지위 추정능력이 높은 것으로 판명되었다. 본 연구에서는 이와 같이 개발된 생태권역별 상수리나무의 지위지수 추정식을 이용하여 적지면적을 산출하고 적지분포를 도해하였다.

• 통상정보연구
• /
• 제16권3호
• /
• pp.121-138
• /
• 2014

본 논문에서는 한 중 무역 현황과 특징을 파악하고 앞으로 한 중 무역의 발전방향을 모색하기 위해 실증분석을 통하여 한 중 무역량에 영향을 미치는 결정요인들의 상관관계를 분석하고자 한다. 2000년부터 2013년까지 14년간의 분기별 데이터를 시용하여 실증분석 방법을 통해 도출된 분석결과는 한 중 양국의 GDP가 높을수록 한 중 무역량에 정(+)의 영향이 유의적인 결과를 나타냈다. 한국의 직접투자가 한 중 무역량에 정(+)의 영향이 있다는 것을 보인 반면에 중국의 직접투자가 한 중 무역량에 부(-)의 영향이 있다는 것을 나타냈다. 한 중 양국의 대외개방도가 높을수록 한 중 무역량의 영향이 유의한다는 것을 보였다. 그리고 한 중 양국의 경제자유도가 높을수록 한 중 무역량에 대해 유의하지 않다는 것을 확인하였다. 연구 시사점은 중국의 대한국 투자분야에 있어 포트폴리오 부동산 투자분야 보다는 그린필드투자(Greenfield)를 적극 유도할 필요가 있으며, 한중 FTA 체결도 보다 신속하게 체결할 필요가 있다고 하겠다.

• 대한인간공학회지
• /
• 제34권4호
• /
• pp.313-326
• /
• 2015

Objective: The objective of this research is to quantitatively analyze muscle activities of arm and shoulder, according to direction in various types of one-handed manual material handling, based on surface electromyography. Background: Workers in industrial sites frequently carry out one-handed manual material handling using arm and shoulder muscles. Therefore, chronic load and accumulated fatigue occur to arm and shoulder muscles, which becomes a main cause of upper arm and shoulder musculoskeletal disorders. The shoulder muscles have widely range of motion, and complex interactions take place among various muscles including rotator cuff muscles. In this regard, research on interactions among should muscles, according to such various dynamic motions, is required. Method: Ten male subjects in their 20s participated in this research. This research considered upward, downward, leftward, rightward, forward and backward directions and fourteen muscles around arm and shoulder (biceps brachii and trapezius, etc.) as independent variables. The mean muscle activity was set as the dependent variable. This research extracted $4^{th}{\sim}7^{th}$ repetition signals according to ten times of repetitive muscle contraction, and analyzed the muscle activity concerned using the envelope detection technique. Results: The mean muscle activity of upward direction was analyzed highly statistically significant. The reason is that the effect of gravity works to arm and shoulder muscles. Also, it is conjectured that deformation of coracoacromial ligament was caused, and its contact pressure increased, due mainly to the shoulder flexion, and therefore load was analyzed high. Muscle activity was analyzed significantly low, according to concentric ballistic motion used in the concentric contraction phase by storing elastic energy in the eccentric contraction phase with a motion to bring the weight to the front of subject's body as to downward, leftward and backward directions. Because, elbow joint's flexion-extension motions mainly occurred, biceps brachii was analyzed high muscle activity as the prime mover. Conclusion: The information on the quantitative load of muscles can be applied to ergonomic work design for one-handed manual material handling to minimize muscle load. Application: This research has effectively identified muscle activity according to dynamic contraction by applying an envelope detection technique. The results can be used for ergonomic work design to minimize muscle load during the one-handed manual material handling, according to each direction. The research results are expected to be used for musculoskeletal disorder prevention and physiotherapy in the rehabilitation medical field, based on the muscle load of arm and shoulder in various directions.