• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권2호
• /
• pp.810-831
• /
• 2019

Recently, continuous dimensional emotion recognition from audiovisual clues has attracted increasing attention in both theory and in practice. The large amount of data involved in the recognition processing decreases the efficiency of most bimodal information fusion algorithms. A novel algorithm, namely the incomplete Cholesky decomposition based kernel cross factor analysis (ICDKCFA), is presented and employed for continuous dimensional audiovisual emotion recognition, in this paper. After the ICDKCFA feature transformation, two basic fusion strategies, namely feature-level fusion and decision-level fusion, are explored to combine the transformed visual and audio features for emotion recognition. Finally, extensive experiments are conducted to evaluate the ICDKCFA approach on the AVEC 2016 Multimodal Affect Recognition Sub-Challenge dataset. The experimental results show that the ICDKCFA method has a higher speed than the original kernel cross factor analysis with the comparable performance. Moreover, the ICDKCFA method achieves a better performance than other common information fusion methods, such as the Canonical correlation analysis, kernel canonical correlation analysis and cross-modal factor analysis based fusion methods.

• 아동학회지
• /
• 제26권2호
• /
• pp.75-89
• /
• 2005

The subjects of this study were 172 5-year-old children enrolled in a kindergarten located in Daejeon city. Instruments were the K-ABC(1997) to test children's intelligence and the TTCT(1999) to test children's creativity, Data were analyzed by Pearson's R and Canonical R. Results indicated that the relation between K-ABC IQ and TTCT CQ was statistically significant(r=.292, p<.000). This result is at variance with threshold theory where the correlation between intelligence and creativity is predicted to be high for children with IQ below 120, but above that level, the correlations are predicted to be much lower. A significant canonical function(Wilk's =.760, p<.000) was found between the sequential processing scale, achievement scale of K-ABC and elaborateness, resistance to premature closure of TTCT.

• 대한가정학회지
• /
• 제49권10호
• /
• pp.103-113
• /
• 2011

This study investigated the structural relationships between attachment, social competence, and emotion regulation. A survey was administered to 233 children of elementary school age(5th-6th grades) in G-city, Korea, using the IPPA-R, the Social Competence Inventory and an Emotion Regulation Scale. The collected data were then analyzed using a Canonical Correlation Analysis. First, the relationship between attachment and social competence was analyzed. The results showed that attachment and social competence have a positively correlated relationship. Peer attachments strongly affect the attributes of social competence when using a canonical variate analysis. Secondly, the relationship between attachment and emotional regulation was analyzed. The results showed that attachment and emotion regulation are also positively correlated. Maternal attachment particularly strongly affected the attributes of emotion regulation. Thirdly, the relationship between social competence and emotional regulation was analyzed. The results showed that social competence and emotional regulation have a positive relationship.

• 한국환경과학회지
• /
• 제11권10호
• /
• pp.1031-1037
• /
• 2002

CCA (Canonical Correspondence Analysis) was used so as to analyze the relation between vegetation and soil environment of Mankyeong river located in Jeollabuk-do. Vegetation survey consulted 1:5,000 topographical map, set up 30 plot and analyzed from June, 2001, to september, 2001. Plant communities of Mankyeong river was investigated by phytosocialogical method. The species composition of plant communities showed high correlation to soil pH, soil organic matter, $P_2$$O_5$, total nitrogen, EC, when they were analyed by CCA. According to the results of CCA hydrophyte communities were distributed in the region that high pH. But the vegetation of disturbed site and wetland plants were distributed in a good nutrients.

• 한국간호교육학회지
• /
• 제3권
• /
• pp.78-98
• /
• 1997

Health education, as the most fundamental concept for national health promotion, alms for developing the self-care ability of the general public. High school days are regarded as the period when most important physical, mental and social developments occur, and most health-related behaviors are formed. School health education is one of the major learning resources influencing health potential in the home and community as well as for the individual student. High school health education in Korea has a fundamental systemic flaw in that health-related subjects are divided and taught under various subjects areas at school. In order to achieve the goal of school health education, it is essential to make a systematic assessment of the learner's concerns connected with his health and life, and the factors affecting them. So far, most of the research projects that had been carried out for improving high school health education were limited in their concerns to a particular aspect of health. Even though some had been done in view of comprehensive school health education, they failed to Include a health assessment of the learner. Therefore, in this study the high school students' concerns related to health and life were investigated in the first place on the basis of the PRECEDE model, developed by Green and others for the purpose of a comprehensive diagnostic research on high school health education. This study was done in two steps : one was the basic study for developing research instrument and the other was the main one. The former was conducted at five high schools in Seoul and Cheongju for 2 months-beginning in March, 1996. The students were asked to respond to questions related to their health and lives in unstructured open-ended question forms. On the basis of analysis of the basic study, the diagnostic instruments for the quality of life, health problems, health behavior and educational factors were constructed to be used for the collection of data for main study. An expert panel and the pilot study were used to improve content validity and reliability of the instruments. The reliability of the instruments was measured at between .7697 and .9611 by the Cronbach $\alpha$. The data for this study were collected from the sample consisted of the junior and senior classes of twenty general and vocational high schools in Seoul and Cheongju for two months period beginning in July, 1996. In analyzing the data, both t-test and $X^2$-test were done by using SAS-$PC^+$ Program to compare data between the sexes of the high school students and the types of high school. A canonical correlation analysis was carried out to determine the relationships among the diagnostic variables, and a multivariate multiple regression analysis was conducted by using LISREL 8.03 to ascertain the influences of variables on the high school students' health and quality of life. The results were as follows : 1) The findings of the hypothesis tests (1) The canonical correlation between the educational diagnosis variables and behavioral, epidemiological, social diagnosis variables was .7221, which was significant at the level of p<.001. (2) The canonical correlation between the educational diagnosis variables and the behavior variables was .6851, which also was significant (p<.001). (3) The canonical correlation between the behavioral diagnosis variables and the epidemiological variables was 4295, which was significant (p<.001). (4) The canonical correlation between the epidemiological diagnosis variables and the social variables was .6005, which was also significant (p<.001). Therefore, the relationship between each diagnosis variable suggested by the PRECEDE model had been experimentally proven to be valid, supporting the conceptual framework of the study as appropriate for assessing the multi-dimensional factors affecting high school students' health and quality of life. Health behavior self-efficacy, the level of parents' interest and knowledge of health, and the level of the perception of school health education, all of which are the educational diagnostic variables, are the most influential variables in students' health and quality of life. In particular, health behavior self-efficacy, a causative factor, was one of the main influential variables in their health and quality of life. Other diagnostic variables suggested in the steps of the PRECEDE model were found to have reciprocal relations rather than a unidirectional causative relationship. The significance of this research is that it has diagnosed the needs of high school health education by the learner-centered assessment of variety of factors related to the health and the life of the students. This research findings suggest an integrated system of school health education to be contrived to enhance the effectiveness of the education by strengthening the influential factors such as self-efficacy to improve the health and quality of the lives of high school students.

• 대한토목학회논문집
• /
• 제26권5D호
• /
• pp.885-894
• /
• 2006

본 논문의 의도는 하이퍼스펙트럴 영상의 다량의 밴드를 사용하면서도 효율적인 분류기법의 개발에 초점을 두고 있다. 본 연구에서는 하이퍼스펙트럴 영상의 분류에 있어 이론적으로 밴드수가 많아질수록 분류정확도가 높을 것이라 예상되는, 다변량 통계분석기법중의 하나인 정준상관분석을 적용한 분류기법을 제안한다. 그리고 기존의 대표적인 전통적 분류기법인 최대 우도분류 방법과 비교한다. 사용되는 하이퍼스펙트럴 영상은 2001년 9월 2일 취득된 EO1-Hyperion 영상이다. 실험을 위한 밴드수는 LANDSAT TM 영상에서 열밴드를 제외한 나머지 데이터의 파장대와 일치하는 부분을 감안하여 30개 밴드로 선정하였다. 지상실제데이터로서 비교기본도를 채택하였다. 이 비교기본도와 시각적으로 윤곽을 비교하고, 중첩분석하여 정확도를 평가하였다. 최대우도분류의 경우 수역 분류를 제외하고는 전혀 분류기법으로서의 역할을 하지 못하는 것으로 판단되며, 수역의 경우도 큰 호수 외에 작은 호수나 골프장내 연못, 부분적으로 물이 존재하는 작은 영역 등은 전혀 분류하지 못하고 있는 것으로 나타났다. 그러나 정준상관분류결과는 비교기본도와 형태적으로 시각적 비교를 해볼 때 골프장잔디를 거의 명확히 분류해 내고 있으며, 도시역에 대해서도 고속도로의 선형 등을 상당히 잘 분류해내고 있음을 알 수 있다. 또한 수역의 경우도 골프장 연못이나 대학교내 연못, 기타지역의 연못, 웅덩이 등 까지도 잘 분류해내고 있음을 확인할 수 있다. 결과적으로 정준상관분석 알고리즘의 개념상 트레이닝 영역 선정시 시행착오를 겪지 않고도 정확한 분류를 할 수 있었다. 또한 분류항목 중에서 잔디와 그 외 식물을 구분해 내는 능력과 수역을 추출해 내는 능력이 최대우도분류기법에 비해 우수하였다. 이상의 결과로 판단해 볼 때 하이퍼스펙트럴영상에 적용되는 정준상관분류기법은 농작물 작황 예측과 지표수 탐사에 매우 유용하리라 판단되며, 나아가서는 분광적 고해상도 영상인 하이퍼스펙트럴 데이터를 이용한 GIS 데이터베이스 구축에 중요한 역할을 할 수 있을 것으로 기대된다.

• 생태와환경
• /
• 제46권4호
• /
• pp.561-569
• /
• 2013

본 연구에서는 내장산국립공원의 산림지역을 대상으로 하여 식생의 구조에 영향을 미치는 입지환경 요소를 평가하고자 한다. 이를 위해서 식생의 구조변수로 엽면적지수(Leaf Area Index, LAI), 흉고직경, 수고로 하고 이에 영향을 주는 입지환경요소로 해발고도, 지형경사, 사면방향, 토양 수분조건, 지형적 위치, 그리고 일사량으로 하여 상호 관계성을 분석하기 위해 정준상관분석 방법을 적용하였다. 입지환경 특성과 식생구조 특성의 단순 상관관계를 살펴보면 전체적으로 상관계수가 0.6 미만으로 낮게 나타나고 상관성 방향을 보면 주로 고도가 높은 능선지역에 수분이 많고 태양복사량이 많을수록 LAI값이 낮아지는 것으로 나타났고 이와 달리 경사가 급하고 동쪽사면에서 북쪽사면으로 갈수록 LAI값은 높아지는 경향을 보였다. 흉고직경은 고도가 높고 경사가 급할수록 작아지는 경향이 나타나고 수분함유량과 태양복사에너지량이 많을수록 흉고직경은 커지는 것으로 나타났다. 수고는 수분이 많고 능선지역에 가까울수록 낮아지는 것으로 나타났다. 이러한 다양한 상관성은 국립공원의 다양한 입지환경 특성이 반영된 결과로 판단되고 식생구조변수는 단일요소가 아닌 복합적인 입지환경 요소에 영향을 받는다는 것을 간접적으로 판단할 수 있는 기준이 될 수 있다. 식생구조에 입지환경요소의 복합적인 영향을 정준상관분석 방법을 수행한 결과 식생구조에 미치는 입지환경의 영향은 일사량이 적고 수분이 적은 고도가 높을수록 흉고직경 및 수고는 작아지고 LAI 커지는 관계성이 나타났다. 비록 산림 식생구조에 미치는 기후를 포함한 다양한 요인에 대한 고려가 필요하지만 산림지역에서 일사량 및 수분조건이 열악한 고지대의 환경에서 적응할 수 있는 식생구조 특성을 통계적으로 유추해 보는데 의미가 있다고 판단된다. 본 연구에서 유도된 결과는 DEM기반의 입지환경 요소는 산림 식생 구조의 다양성을 평가하는 데 유용한 방법으로 적용될 수 있고, 국립공원에서 산림식생구조를 주기적으로 관찰하고 효과적인 공원관리대책을 수립하는데 기초자료로 활용할 수 있을 것으로 판단된다.

• 대한간호학회지
• /
• 제41권3호
• /
• pp.354-363
• /
• 2011

Purpose: The purpose of this cross-sectional study was to examine the relationship between characteristics of severe ALS patient-caregiver couples and health related quality of life (HRQoL) in family caregivers. Methods: The participants in this study were 89 pairs of ALS patients using ventilators and a family caregiver. The characteristics of the ALS patients and caregivers, Korean-Amyotrophic Lateral Sclerosis Functional Rating Scale-Revised, Zarit Burden Interview and SF-36 were measured in this study. The data were collected from August 2008 to April 2009. Descriptive statistics, Pearson correlation coefficients, and canonical correlation were used for data analysis. Results: The physical component summary and mental component summary of the HRQoL score for family caregivers were $147.49{\pm}31.63$ and $129.09{\pm}35.83$, respectively. HRQoL for caregivers was related to characteristics of the ALS patient-caregiver couples, such as patient's gender, caregiver's age, gender, marital status, daily time spent in caregiving and burden with one significant canonical variable. The significant variate showed that the lower the age, the time spent in caregiving and the burden of caregivers, the higher the HRQoL of caregivers. Conclusion: The support systems for caregivers considering caregiver characteristics such as demographics and burden should be implemented to improve the HRQoL of caregivers.

• Child Health Nursing Research
• /
• 제17권2호
• /
• pp.127-135
• /
• 2011

Purpose: The purpose of this study was to examine the relationship of mothers' recognition of Attention deficit hyperactivity disorder (ADHD), parenting stress and family support in mothers of children diagnosed with ADHD. Methods: The data were collected from December 24, 2009 to July 23, 2010. The participants were 141 mothers of children diagnosed with ADHD and who lived in D city. Descriptive statistics, t-test, ANOVA, Pearson correlation coefficients with SPSS 17.0 were used to analyze the data. To determine the most effective variable (mothers' recognition, parenting stress and family support) to predict parental stress, data were analyzed using canonical correlation with SAS 9.1 TS. Results: Canonical correlation analysis revealed that professional and teachers' roles of the 3 sub-domains of mothers' recognition were the most outstanding variables in predicting parenting stress. Parental distress/parent-child dysfunctional interaction of parenting stress was the most outstanding variable in predicting both mothers' recognition and family support. Spouse cooperation in family support was the most outstanding variable in predicting parenting stress. Conclusion: The results suggest that strengthening parents' recognition of ADHD and supporting interventions are important for reducing parenting stress. Professional and teachers' roles, parental distress/parent-child dysfunctional interaction and spouse cooperation should be included in the contents of structured programs.

• 한국환경생태학회지
• /
• 제32권5호
• /
• pp.497-506
• /
• 2018

본 연구는 매토종자를 활용한 소나무림 생태복원 시 적정 매토종자 최소생존가능개체군의 크기를 확인하기 위해 2010년 8월부터 2011년 11월까지 매토종자 발아실험을 실시하였다. 상관관계분석 결과, 면적과 초본 종수가 0.686으로 가장 높게 나타났다. 도서생물지리학의 이론에 따라 최소생존가능개체군에 적용하여 면적과 유의성이 있는 네 변수에 대해 회귀분석을 한 결과 목본 종수, 목본 개체수, 초본 종수 및 초본 개체수 모두 5% 이내에서 유의하며, 회귀함수의 설명력은 58.3%였다. 종수와 개체수간의 정준상관분석 결과 한 개의 함수가 유의하였으며 함수의 설명력은 82.4%이고, 유의수준 1% 이내에서 정준함수의 정준근 모두 유의한 것으로 판단된다. 연구 결과, 소나무림 매토종자 최소생존가능 개체군의 크기는 면적 $64m^2$이상, 종수 21종 이상, 개체수 120개체 이상을 권장하며 초본 종수의 영향에 따라 매토종자 군집의 크기가 결정되는 것으로 나타났다. 따라서, 매토종자를 활용한 소나무림 생태복원의 적용 시 초본 종에 대한 고려가 필요하다.