• 대한교통학회지
• /
• 제23권3호
• /
• pp.73-84
• /
• 2005

도로기하구조 설계기준은 운전자의 안전성 향상과 원활한 주행능력을 제공할 수 있도록 마련되었으나 지형조건이 열악한 우리나라의 경우 부득이 설계기준이 의도하는 방향과 운전자가 인지하는 특성에 차이를 보임으로써 안전성에 문제점이 있는 경우가 많다. 이러한 교통사고의 위험성을 감소시키기 위한 대책 중에 하나로 도로안전시설물을 설치하지만 설치한 안전시설물에 대한 평가는 제대로 이루어지지 않고 있으며, 특히 도로환경적 요인과 인적요인과의 관련성에 대한 연구는 상당히 미흡한 실정이다. 본 연구에서는 도로 이용자들이 종단선형구간에서 느끼는 도로안전성 요인을 파악하기 위하여 차량시뮬레이터(Driving Simulator)와 3D 그래픽 화면으로 종단선형구간을 재현하고 도로안전시설물의 설치 유, 무에 따라 운전자가 느끼는 안전성의 변화와 도로 이미지를 조사하였다. 조사된 데이터를 바탕으로 수량화 제II류에 의한 정준상관분석을 실시하여 도로 안전성의 영향 요인을 분석하였으며 수량화제 I류에 의한 모형화를 실시하여 안전성을 예측한 결과 도로안전시설물과 운전자가 느끼는 안전성과의 관계를 명확히 규명할 수 있었다.

• Korean Chemical Engineering Research
• /
• 제45권1호
• /
• pp.87-92
• /
• 2007

최근 공정의 이상을 감지하고 진단하기 위한 공정 모니터링 시스템의 개발이 공정 시스템 분야에서 많은 주목을 받고 있다. 공정으로부터 얻어지는 데이터는 공정의 특성에 대한 유용한 정보를 제공하고 이는 공정의 모델링과 모니터링 그리고 제어에 사용된다. 현대의 화학 및 환경 공정은 고차원적인 특성과 변수간의 강한 상관관계와 동특성 그리고 비선형적 특성을 가지고 있어 모델 기반 접근을 통해 공정을 분석하는 것을 쉽지 않다. 이러한 모델 기반 접근의 한계를 극복하기 위해 많은 시스템 엔지니어와 연구자들이 주성분 분석법(principal component analysis, PCA) 또는 부분 최소 자승법(partial least squares, PLS)과 같은 다변량 분석을 접목한 통계 기반 접근법에 초점을 맞추고 있다. 또한 동특성, 비선형성 등과 같은 특성을 가진 공정에 적용하기 위해 많은 다변량 분석법들이 보완되었다. 여기에서는 동적 주성분 분석법(dynamic PCA)과 케노니컬 변수 분석법(canonical variate analysis)을 이용한 결측 데이터의 예측법과 공정 변수의 복원을 통한 센서 오작동의 판별법에 대해 언급해 보고자 한다.

• 대한간호학회지
• /
• 제16권2호
• /
• pp.63-69
• /
• 1986

The main purpose of this study was to find out variables relevant to self-care activities of physically disabled person. The subjects of this analysis were 1277 person which is between 15~64 years, the data came from the 1985 National Interview Survey on Disabled Person in Korea. For this analysis, Breakdown, Oneway and Discriminant Analysis were used. The finding of the analysis can be summarized as follows: First, the mean of self care activities was 2.57 (SD: 0.69, range: 1-3). The relevance for the self-care activities by several variables is as follows. 1. The relevance for the self-care activities by socioeconomic status is significant at age, education level, occupation of household members variables. Especially, in the case of high age, low education level, the self-care activities are shown low score. 2. The relevance by impairment characteristics is shown high significance at all input variables. When disabled person have double impairment. paralysis, late occurance age, and is due to diseases the self-care activities score is lowered. 3. The relevance by health care services variables. is not shown significant at all input variables. Second, the relevance for social activities by sev-eral variables was conducted by discriminant analysis. The relative importance of social activities discriminant function is 0.344 of eigenvalue. The-canonical correlation between the social activities discriminant function and 9 dummy variables is 0.51, total variance of dummy variables for social activities is shown 26 persent. The self-care activities variable represents the highest contribution of its associated variable to the function (canonical coefficient: -.56). The occurance age, the occupation of household members, the education level variables are shown comparatively high contribution to the function. To sum up, this analysis suggests that the self-care activities variable is the highest contributed to the social activities. In relation to self-care concept, this finding will be useful in rehabilitation nursing care.

• 응용통계연구
• /
• 제37권2호
• /
• pp.157-176
• /
• 2024

생명현상의 복잡한 시스템에 대한 이해를 위한 융합분석의 중요성이 점점 커지고 있다. 하나의 연구대상을 다양한 관점에서 관찰하여 얻게 되는 여러 데이터의 융합분석은 통해 좀 더 대상에 대한 깊은 이해를 가능하게 한다. 본 연구에서는 그중에서도 특히 하나의 샘플에서 두개의 고차원 데이터가 생성된 경우 다룰 수 있는 분석인 공관성분석과 정준상관분석을 비교하였다. 정준상관분석의 경우 고차원 데이터를 다룰 수 없는 단점이 있기에, 해당 문제를 극복하기 위하여 능형상수를 이용하는 방법(CCA-ridge)과 각 데이터의 공분산행렬을 항등행렬로 가정하여 벌점화 특이값분해를 이용한 방법(CCA-PMD) 두 가지를 고려하였으며 각 방법을 NCI60 세포주 패널에서 얻은 RNA 시퀀싱 데이터와 단백질 시퀀싱 데이터 분석에 적용하였다. 그 결과 정준상관분석의 경우 두 정준변수간의 상관관계에 좀 더 집중하는 반면 공관성분석은 각 데이터의 선형조합간의 상관관계뿐 아니라 각 선형조합의 변동성을 함께 고려함을 확인할 수 있었다. 또한 공관성분석의 경우 여러가지의 가중치행렬을 고려하여 그 결과값을 비교하고 중요 시사점을 도출하였다.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 1999년도 Proceedings of International Symposium on Remote Sensing
• /
• pp.140-143
• /
• 1999

SSH(Sea Surface Height) from TOPEX/Poseidon and SST data are analyzed to estimate characteristics of annual and inter-annual variations in the East Asian seas(110E - l80E, 20N - 50N) from November 1992 to May 1998. In EOF analysis of SSH and SST, 57% and 97% of the variance are represented by the first two modes. The first mode of SSH and SST shows strong annual variations expected for steric changes. The second mode of SSH shows a long-term variation, but that of SST shows 3 - 4month offset from annual variation. In the EOF analysis of the SSH and SST, 61% and 54% of the total variance are represented by the first three modes. The first mode represented a long-term variation, and the third mode reflected the bi-ennial variation. The first three mode were not strongly correlated with ENSO. We further apply Canonical Correlation Analysis(CCA) to find the dominant correlation patterns with ENSO.

• Child Health Nursing Research
• /
• 제10권2호
• /
• pp.205-216
• /
• 2004

Purpose: The purpose was to investigate the factors influencing alcohol consuming behavior of the female university students to provide the basic data for a nursing intervention program to improve health and prohibit of alcohol consuming behavior. Method: The subjects were 134 female university students, studying at K university and K educational university in Chungnam province from April 1 to 20, 2002. The instruments were the alcohol consuming behavior scale, that is drinking intensity score, and alcohol problem developed by Shin(1998), refusal self-efficacy scale by Aas et. al.(1995), alcohol expectancy by Goldman at. al.(1989), TPQ scale by Cloninger(1991), depression scale modified by Shin(1977) using Zung(1965)' tool, family cohesion scale by Olson at. al.(1983). The data were analyzed using descriptive statistics, Pearson correlation coefficient, canonical correlation and stepwise multiple regression. Result: The most powerful predictor of alcohol consuming behavior was influence of friends-drinking(21.0％). A combination of economic status(8.0％), and personality of novelty seeking(3.0％), accounted for 32.0％ of the variance in alcohol consuming behavior. Conclusion: I recommended that economic status, influence of friends drinking frequently, and personality of novelty seeking were contained of the developing nursing intervention program for decreasing the alcohol consuming behavior in female university students.

• 가정과삶의질연구
• /
• 제13권4호
• /
• pp.55-70
• /
• 1995

The purpose of this study were to develop 「The Inventory of Children's Social Support」 to investigate the sex difference of the perceived social support and social support satisfaction level and to identify the relation between children's social support system and children's social competence. The subjects were 213 5th, 6th children and their 6 teachers of elementary school at Kwang Ju. The data were analyized by frequency percentile factor analysis t-test and canonical correlation analysis, The results were as follows ; 1) The reliability and validity of 「The Inventory of Children's Social Support」was confirmed. 2) Sex difference of perceived social support and social support satisfaction level were partially found 3) There were significant relations between social support system and social competence of children. In conclusion children's social support system can be recognized as significant variable in predicting the social competence of children.

• 아동학회지
• /
• 제28권3호
• /
• pp.61-75
• /
• 2007

This study defined awareness of professionalism and job stress in 353 child-care teachers of children with special needs and analyzed the relative contribution of professional awareness to job stress. Teachers showed high awareness of 'ethics' and 'enthusiasm enhancing professionalism', and this awareness correlated highly with professionalism in the 'specialized knowledge and skills' sub-factors. Teachers showed high levels of stress in the 'general duty' and the 'guidance of children with special needs' categories, and this correlated highly with stress caused by 'child-care center management.' The results of canonical correlation analysis indicated that of professionalism sub-factors only the awareness of 'social services' contributed positively to job stress.

• 대한간호학회지
• /
• 제27권1호
• /
• pp.189-201
• /
• 1997

This study was designed to test the relationship between effectiveness of nursing organizations and structural and managerial variables of nursing organizations that are described in the Robbins Organizational theory model. The data were collected through self reported questionnaires from 605 nurses working in, and 782 patients hospitalized in, five tertiary hospitals in Seoul. Results showed that according to MANOVA there was a significant difference in nurses job satisfaction and patient satisfaction among the five hospitals. According to cluster analysis of the structural and managerial variables of nursing organizations, the five hospitals were divided into two clusters and there was no significant difference in nurses job satisfaction or patient satisfaction between the two clusters. According to canonical correlation analysis the formalization and centralization of structural variables were shown to be predicting variables for nurses job satisfaction, and the managing job design and managing Change of managerial variables were shown to be predicting variables for nurses job satisfaction.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제7권4호
• /
• pp.819-833
• /
• 2013

Computing global features based on local features using a bio-inspired framework has shown promising performance. However, for some tough applications with large intra-class variances, a single local feature is inadequate to represent all the attributes of the images. To integrate the complementary abilities of multiple local features, in this paper we have extended the efficacy of the bio-inspired framework, HMAX, to adapt heterogeneous features for global feature extraction. Given multiple global features, we propose an approach, designated as parameterized metric learning, for high dimensional feature fusion. The fusion parameters are solved by maximizing the canonical correlation with respect to the parameters. Experimental results show that our method achieves significant improvements over the benchmark bio-inspired framework, HMAX, and other related methods on the Caltech dataset, under varying numbers of training samples and feature elements.