• Journal of the Korean Data and Information Science Society
• /
• 제28권6호
• /
• pp.1291-1300
• /
• 2017

고차원 자료에서는 관측값의 개수보다 변수의 개수가 과다하게 많은 것이 특징이다. 그러므로 회귀 계수 추정에 있어 관측값의 영향이 매우 클 수 있다. Jang과 Anserson-Cook (2017)은 라쏘추정량 사용시 영향점의 영향을 평가할 수 있는 라쏘 영향그림을 제안하였다. 본 연구에서는 고차원 자료에서 영향점을 평가하기 위한 그래픽 방법들로서 라쏘 영향그림 뿐만 아니라 라쏘 변수선택 순위그림, 삼차원 라쏘 영향그림을 제안하였다. 실세 두 가지 고차원 자료 예들에 영향점들을 찾기 위한 회귀진단 수단으로서 세가지 그래픽 방법들을 사용하여 본 결과 영향점들을 효과적으로 찾아낼 수 있었다.

• 한국콘텐츠학회논문지
• /
• 제16권4호
• /
• pp.455-465
• /
• 2016

본 연구는 트위터 유력자가 이용자의 온-오프라인 의견표명 의도에 미치는 영향을 살펴보고자 하였다. 침묵의 나선 이론과 수정 행동 가설에 근거하여 상반된 가설을 도출하였고, 어느 가설이 더 타당한지를 검증해보기 위해 실험연구를 진행하였다. 실험참가자들은 입양특례법 개정안에 대한 공인 유력자의 찬성 트윗, 공인 유력자의 반대 트윗, 일반인 유력자의 찬성 트윗, 일반인 유력자의 반대 트윗이라는 4개의 실험 조건에 배치되었다. 입양특례법 개정안에 대한 실험참가자의 의견을 바탕으로, 실험집단은 공인 유력자와의 의견일치 집단, 공인 유력자와의 의견불일치 집단, 일반 유력자와의 의견일치 집단, 일반 유력자와의 의견불일치 집단으로 구분되었다. 분석결과, 유력자와의 의견일치여부는 이용자의 의견표명 의도에 영향을 주지 않았다. 그러나 유력자 유형에 따른 상호작용 효과가 발견되었는데, 일반인 유력자와의 의견불일치가 이용자들의 온-오프라인 의견표명 의지를 높이는 것으로 나타났다. 본 연구결과는 트위터 유력자가 이용자들을 침묵시키는 것이 아니라 오히려 의견표명을 동기화시킬 가능성이 있음을 보여준다.

• 디지털산업정보학회논문지
• /
• 제10권4호
• /
• pp.211-224
• /
• 2014

This study is aimed at considering the relationship between the influential factor in Smartphone purchase and consumer awareness. For this purpose, this study set up the influential factors in Smartphone purchase as advertising effect, brand effect and usability effect and corporate image effect, and considered the relationship between these influential factors and consumer awareness empirically. The research results can be summarized as following 3 parts: First, the four factors suggested as the influential factor in Smartphone purchase were verified to have a significant influence on customer satisfaction and customer loyalty, and thus, it is thought that the bigger the influence of these factors, the more customer satisfaction and customer loyalty will increase. Second, among the factors suggested as influential factors in Smartphone purchase, the biggest factor in influencing customer satisfaction was found to be 'Usability Effect' while the biggest factor in influencing customer loyalty was revealed to be 'Brand Effect.' As a result, this study could learn that it's necessary that 'Usability Effect' should be considered more preferentially than other factors in order to enhance customer satisfaction while 'Brand Effect' should be preferentially considered in orderto raise customer loyalty. Third, it can be judged that in the relationship between customer satisfaction and customer loyalty, the bigger the customer satisfaction, the higher the customer loyalty. It is hoped that such research results could be applied as useful materials in time of establishing the marketing strategy. Nevertheless, this paper makes it clear that due to the constraints in the number of sample areas and samples, there is a limit in objectifying the research results into all areas.

• 수산경영론집
• /
• 제36권2호
• /
• pp.53-69
• /
• 2005

The recent emergence of electronic commerce could provide different opportunities to small firms in overcoming part of their technological, environmental, organizational, and managerial inadequacies. However, recent research provided a gloomy picture about electronic commerce uptake and use in small firms. Until now few small flrms adopted electronic commerce. This phenomenon gives us several implications. One of the implications is a need to generate more research to small firms' uptake and use of electronic commerce. So, this research aims at what is determinants in small firms' adoption of electronic commerce, especially business - to - business(B2B). Several previous studies were identified various influential variables in adoption of small firms' electronic commerce. To some extent, by industries these variables may be different in influencing B2B adoption of small firms. This study selected fishery industry's wholesale firms as survey domain. Then, It selected some variables from previous studios and group them in several factors for consistently comparing variable's influential power. Through hierarchical influencing model of B2B adoption, managers who worked for fishery wholesale firms were surveyed. Among the first level's factors of the model, business factor was identified as the statistically most influential factor in B2B adoption. Technological factor and environmental factor at the frist level were identified as relatively less influential factors. Among the second level's factors of the model, it was statistically significant only that technological usefulness factor was more influential than technological burden factor in B2B adoption. And among the third level's factors of the model, it was identified that formal plan and task team for B2B, top management support, and mutual beliefs of inter - firms were statistically more influential than related variables in B2B adoption. These results give us some implications. First, fishery wholesale firms recognized B2B as a new business paradigm, and technological problems as not obstacles in adopting B2B. Thus it would be possible to activate the fishery B2B if they were guided to getting B2B benefits and B2B's relative advantages. Second, they considered the power of influential factors might be different by B2B stages. That is, top management support was more important in the B2B introduction stage, and formal plan and task team for B2B and mutual beliefs of inter - firms were more important in the B2B operation stage.

• Journal of the Korean Statistical Society
• /
• 제20권2호
• /
• pp.177-190
• /
• 1991

In this paper, we discuss and review various measures which have been presented for studying outliers. high-leverage points, and influential observations when principal component regression is adopted. We suggest several diagnostics measures when principal component regression is used. A numerical example is illustrated. Some individual data points may be flagged as outliers, high-leverage point, or influential points.

• Communications for Statistical Applications and Methods
• /
• 제6권1호
• /
• pp.43-52
• /
• 1999

The local influence method is adapted to quadratic discriminant analysis for the identification of influential observations affecting the estimation of probability density function probabilities and log odds. The method allows a simultaneous perturbation on all observations so that it can identify multiple influential observations. The proposed method is applied to a real data set and satisfactory result is obtained.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권6호
• /
• pp.2979-2995
• /
• 2017

In recent years, there has been an increasing number of studies focused on identifying a set of spreaders to maximize the influence of spreading in complex networks. Although the k-core decomposition can effectively identify the single most influential spreader, selecting a group of nodes that has the largest k-core value as the seeds cannot increase the performance of the influence maximization because the propagation sphere of this group of nodes is overlapped. To overcome this limitation, we propose a neighborhood coreness cover and discount heuristic algorithm named "NCCDH" to identify a set of influential and decentralized seeds. Using this method, a node in the high-order shell with the largest neighborhood coreness and an uncovered status will be selected as the seed in each turn. In addition, the neighbors within the same shell layer of this seed will be covered, and the neighborhood coreness of the neighbors outside the shell layer will be discounted in the subsequent round. The experimental results show that with increases in the spreading probability, the NCCDH outperforms other algorithms in terms of the affected scale and spreading speed under the Susceptible-Infected-Recovered (SIR) and Susceptible-Infected (SI) models. Furthermore, this approach has a superior running time.

• 응용통계연구
• /
• 제31권1호
• /
• pp.97-108
• /
• 2018

회귀모형을 이용하여 자료를 분석하는 경우 이상점이나 영향점의 유무를 검정하는 회귀진단기법은 모형의 적합성을 체크하기 위한 필수적인 도구이다. 이러한 이상점이나 영향점이 존재하는 경우 회귀분석의 결과가 왜곡되어 해석이 된다. Jang과 Anderson-Cook (Quality and Reliability Engineering International, 30, 1409-1425, 2014)은 불꽃그림이란 이름을 붙인 그래픽 방법를 제시하였는데 관측값에 부여된 가중치를 1에서 0으로 변화함에 따라 이상점이나 영향점이 회귀계수 및 잔차제곱합에 어떠한 영향을 미치는지 살펴 보았다. 본 연구에서는 다반응 반응표면분석에서 이러한 불꽃그림을 적용하여 보고자 한다.

• 대한기계학회논문집B
• /
• 제27권6호
• /
• pp.766-772
• /
• 2003

The pseudospectral method for stability analysis was used to find the most influential disturbance mode for transition of plane channel flows and Blasius flow at their critical Reynolds numbers. A number of various oblique disturbance waves were investigated for their pseudospectra and resolvent norm contours in each flow, and an exhaustive search method was employed to find the disturbing waves to which the flows become most unstable. In plane Poiseuille flow an oblique disturbance with a wavelength of 3.59h (where h is the half channel width) at an angle $28.7^{\circ}$ was found to be the most influential for the flow transition to turbulence, and in plane Couette flow it is an oblique wave with a wavelength of 3.49h at an angle of $19.4^{\circ}$. But in Blasius flow it was found that the most influential mode is a normal wave with a wavelength of $3.44{\delta}_{999}$. These results imply that the most influential disturbance mode is closely related to the fundamental acoustic wave with a certain shear sheltering in the respective flow geometry.

• 한국생활과학회지
• /
• 제24권5호
• /
• pp.701-712
• /
• 2015

This study analyses "Survey of Health and Behavior of Children and Adolescents" which is conducted by National Youth Policy Institute that investigated influential factors in difference between actual obesity and perceptional obesity. GLM (Generalized Linear Mode) was carried out to estimate the effects on the eating habits and physical activities and multinomial logit analysis was carried out to examine the influential factors in difference between actual, obesity and perceptional obesity. The results of this study are shown as follows. First 72.0% of total respondents showed the congruence between actual obesity and perceptional obesity and 24.6% of total respondents perceived them fatter than actual. The remaining 3.5% perceived them thinner than actual. Second, the main effects of actual obesity and perceptional obesity on eating habits were statistically significant, but interaction effect between them was not statistically significant. The main effects of actual obesity and perceptional obesity on physical activities and interaction effect between them were all statistically significant. Third, the influential factors in difference between actual obesity and perceptional obesity were gender, age, school record, education of parents, subjective economic status, and living together with family.