• 산업공학
• /
• 제12권4호
• /
• pp.551-556
• /
• 1999

In this paper, we compare the performance of three data mining classification algorithms(neural network, decision tree, logistic regression) in consideration of various characteristics of categorical input and output data. $2^{4-1}$. 3 fractional factorial design is used to simulate the comparison situation where factors used are (1) the categorical ratio of input variables, (2) the complexity of functional relationship between the output and input variables, (3) the size of randomness in the relationship, (4) the categorical ratio of an output variable, and (5) the classification algorithm. Experimental study results indicate the following: decision tree performs better than the others when the relationship between output and input variables is simple while logistic regression is better when the other way is around; and neural network appears a better choice than the others when the randomness in the relationship is relatively large. We also use Taguchi design to improve the practicality of our study results by letting the relationship between the output and input variables as a noise factor. As a result, the classification accuracy of neural network and decision tree turns out to be higher than that of logistic regression, when the categorical proportion of the output variable is even.

• 보건교육건강증진학회지
• /
• 제6권2호
• /
• pp.4-22
• /
• 1989

This study aims to explore the relationship between housewives' health KAP level and the physical health of families. The data used in this study are obtained from the Last Evaluation Program of the National Note for Health surveyed in July, 1989. The respondents for this study are 770 housewives residing in Chunan and Buan. The reason why this study focus on KAP level of housewives is to find out whether a housewife as a emotional supporter contributes to the physical health of her families. A housewife gives her families emotional satisfaction insteade of economic satisfaction. So she has the most interest in family health among the members of her family. Therefore, housewife's KAP level will influence her family health. The independent variables chosen for the analysis are the general characteristic variables and KAP level. And the dependent variable is the physical health of families which excluded psychosocial one. This level of family health includes weight for family health and seriouseness of disease. The result of this study was summarized as follows. (1) KAP level was significantly correlated with variables which have mainly the socioeconomic characteristics. The variables were: area of residence, education level occupation, self assessment on wealth, and exposure to mass communication. (2) In the analysis of relationship between the general characteristic variables and family health, family health was significantly correlated with almost all variables. The variables were: Presence of the aged, Health status, Experience in disease, Self assessment on health, No. of families, Occupation, Education level, Self assessment on wealth, Concern on health, and Exposure to mass communication. (3) In the analysis of relationship between family health and KAP level, family health was not significantly correlated with KAP level but. (4) Also in the stepwise regression analysis, the general variables account for about 32.1 percent of the variance in the dependent variable, family health. The variable with the greatest explanatory power was presence of the aged. On the contrary, KAP level explain about 0.4 percent of the variance in the dependent variable. In sum, the study shows that housewives' health KAP level has relatively weak relationship with the physical health of families

• 아태비즈니스연구
• /
• 제15권1호
• /
• pp.209-223
• /
• 2024

Purpose - This article analyzes the impact of macroeconomic variables of the United States, China, and Korea on KOSPI and VKOSPI, in that United States and China have a great influence on Korea, having an export-driven economy. Design/methodology/approach - The influence of US, China, and Korea interest rates, industrial production index, consumer price index, US employment index, Chinese real estate index, and Korea's foreign exchange reserves on KOSPI and VKOSPI is analyzed on monthly basis from Jan 2012 to Aug 2023, using multifactor model. Findings - The KOSPI showed a positive relationship with the U.S. industrial production index and Korea's foreign exchange reserves, and a negative relationship with the U.S. employment index and Chinese real estate index. The VKOSPI showed a positive relationship with the Chinese consumer price index, and a negative relationship with the U.S. interest rates, and Korean foreign exchange reserves. Next, dividing the analysis into two periods with the Covid crisis and the analysis by country, the impact of US macroeconomic variables on KOSPI was greater than Chinese ones and the impact of Chinese macroeconomic variables on VKOSPI was greater than US ones. The result of the forward predictive failure test confirmed that it was appropriate to divide the period into two periods with economic event, the Covid Crisis. After the Covid crisis, the impact of macroeconomic variables on KOSPI and VKOSPI increased. This reflects the financial market co-movements due to governments' policy coordination and central bank liquidity supply to overcome the crisis in the pandemic situation. Research implications or Originality - This study is meaningful in that it analyzed the effects of macroeconomic variables on KOSPI and VKOSPI simultaneously. In addition, the leverage effect can also be confirmed through the relationship between macroeconomic variables and KOSPI and VKOSPI. This article examined the fundamental changes in the Korean and global financial markets following the shock of Corona by applying this research model before and after Covid crisis.

• 복식문화연구
• /
• 제16권6호
• /
• pp.1126-1141
• /
• 2008

The purposes of this study identify how attributes of the website impact on consumer attitude toward the website. For this purpose, the study tested covariance structural model which set relationships among independent variables (interactivity and shopping value), meditated variables(relationship commitment), and dependent variables(attitude toward website). The data were collected from a sample of 239 internet shopper of college female students. The covariance structural model and research hypothesis analyzed by using SPSS 16.0 and AMOS 5.0 program. The results are as follows: First, the structural model is accepted significantly. Second, interactivity and shopping value of website have a positive influence on relationship commitment. Third, interactivity, shopping value of website, and relationship commitment have a positive impact on attitude toward the website. Forth, even if shopping value has not a positive influence on attitude toward the website, it was found to have a significant effect on attitude toward the website through the relationship commitment.

• 농촌지도와개발
• /
• 제17권2호
• /
• pp.305-326
• /
• 2010

This study aimed to investigate couple relationship and related variables of women immigrants. Especially, the study focused on curve term between couple relationship and related variables. The data were collected from a sample of 346 women immigrants living in S city, Chungnam province. First, the study indicated that there was some curve relationship between couple satisfaction and period of residence in Korea. Namely, the couple satisfaction of women immigrants was likely to decrease at the early period of residence, but the longer the period of residence was, the higher the couple satisfaction was. Second, the result revealed that there was no significant relationship between couple satisfaction and level of Korean proficiency. Third, there was some curve relationship between couple satisfaction and satisfaction of life. The couple satisfaction was likely to increase in a shape of curve according to the increase of life satisfaction.

• 한국의류학회지
• /
• 제33권9호
• /
• pp.1472-1482
• /
• 2009

This study identifies how website attributes impact consumer loyalty to websites. For this purpose, the study tested covariance of a structural model which set relationships among independent variables (trust and site stickiness), meditated variables (relationship commitment), and dependent variables (website loyalty). The data were collected from a sample of 239 female college internet student shoppers. The covariance structural model and research hypothesis are analyzed using SPSS 16.0 and AMOS 5.0 program. The results are as follows: First, the structural model is significantly accepted. Second, trust and stickiness have a positive influence on the relationship commitment. Third, trust, stickiness, and relationship commitment have a positive impact on website loyalty. Forth, even if stickiness has no positive influence on website loyalty, it has a significant effect on website loyalty through the relationship commitment.

• 대한가정학회지
• /
• 제44권2호
• /
• pp.137-152
• /
• 2006

The major purpose of this study was to investigate the causal effect of the primary triad experiences on dating relationship satisfaction through self-esteem among college students. The primary triad experiences were composed of the parents' differentiation level of couple relationship and the parent-child relationship including family rule, triangulation and communication. Dating relationship satisfaction included such subscales as problem-solving communication, global satisfaction, affection communication and time together. The study data were collected from 382 college students by using self-administered questionnaire method. The results showed that parents' differentiation was positively correlated with self-esteem and all components of dating relationship satisfaction and that family rule was negatively associated only with problem-solving communication. Triangulation was negatively related with most components of dating relationship satisfaction and self-esteem, while parent-child communication showed positive correlations with the same variables. Path analysis results showed that self-esteem mediated the effect of most variables of primary triad experiences on dating relationship satisfaction. In addition, variables of the primary triad experience had direct influences on most components of dating relationship satisfaction. Self-esteem was found to be the most powerful variable influencing dating relationship satisfaction. The findings of the study were discussed in terms of pre-marital education and counseling.

• 통상정보연구
• /
• 제10권3호
• /
• pp.117-140
• /
• 2008

The purposes of the study are to measure the relative importance of independent variables, to analyze relation between user satisfaction and use, and to improve the level of user satisfaction in using e-Trade system(uTradeHub). The study develops the independent variables in three dimensions(i.e the quality of system, quality of information, and quality of service), estimates the level of satisfaction on a par with these variables, and analyzes the causal relation of these variables. The major findings of the study are as follows. First, there is a significant casual relationship between the quality of system and user satisfaction, thus the hypothesis is accepted. Second, there is a significant casual relationship between the quality of information and user satisfaction, thus the hypothesis is accepted. Third, there is a significant casual relationship between the quality of service and user satisfaction, thus the hypothesis is accepted. Final, there is a significant casual relationship between the user satisfaction and use, thus the hypothesis is accepted.

• 한국가정과학회지
• /
• 제3권2호
• /
• pp.1-12
• /
• 2000

The purpose of this research was to investigate the relationship of family related variables with adolescents'delinquent behaviors. Family related variables were consisted of (a) family's general characteristics (j.e., occupation, income, etc.) and (b) family relationship charateristics (j.e.. parents'child rearing attitude, parents' marital relationship, and communication level between parents and adolescent children) . Adolescents'delinquent behaviors were also analyzed according to adolescents'general charateristics such as sex, grade, birth order, school achievement level, and religion. The data were collected from 577 middle and high school students in Chonju by using self-administered questionnaire method. Results showed that there were significant differences in the frequency level of adolescents'delinquent behaviors across adolescents'sex, grade, birth order, school achievement level, and religion, and across mother's occupation. and subjective living level. The delinquent behaviors were also negatively correlated with most of family relationship characteristics. Results of multiple regression analyses indicated that communication of adolescent children with their father and mother's child rearing attitude were most significant variables influencing adolescents'delinquent behaviors.

• 한국콘텐츠학회논문지
• /
• 제16권7호
• /
• pp.135-145
• /
• 2016

본 연구의 목적은 상업계열 고등학교 학생들의 학교생활만족도에 영향을 미치는 생태체계 변인을 분석하기 위한 것이다. 연구대상은 경기도 소재 상업계열 고등학교 2학년생 312명이며, 설문조사로 수집된 자료의 분석은 SPSS Win 22.0을 이용하여, 기술통계, 상관관계 분석, 위계적 회귀분석과 단계적 회귀분석을 실시하였다. 본 연구의 독립변인은 총 21개로, 자기유능감, 대인관계기술, 내적통제감, 정서인식, 정서표현, 감정이입, 정서조절, 정서활용의 8개 유기체 변인, 학교변인(교사와의 관계, 수업과 학습, 교실분위기, 학교복지시설, 학교주변환경), 친구변인(교우관계), 가정변인(가족기능성, 부의 학력, 경제적 수준)의 9개 미시체계 변인, 부모-교사 관계, 부모-자녀의 친구관계의 2개 중간체계변인, 한국교육제도에 대한 만족도의 1개 외체계 변인, 학벌지상주의의 지지정도의 1개 거시체계변인으로 구성하였다. 분석결과는 교사와의 관계, 한국교육제도에 대한 만족도, 교우관계, 정서활용, 부모-교사 관계 순위로 5개 변인이 유의하였다($R^2=45.3%$). 생태체계 변인 중 미시체계 변인의 영향력은 35.7%로 가장 큰 것으로 밝혀졌다. 본 연구는 생태학적 관점에서 상업계열 고등학생의 학교생활만족도를 분석한 것에서 의의가 있다.