• Journal of the Korean Data and Information Science Society
• /
• 제23권1호
• /
• pp.53-61
• /
• 2012

많은 조사에서 표본 크기는 유의수준 또는 제1종 오류확률만을 고려하여 결정하였으나 최근에는 다양한 분야를 중심으로 제1종 오류확률뿐만 아니라 제2종 오류확률 또는 검정력을 함께 고려하여 표본 크기를 결정하는 경우가 늘어나고 있다. 이런 경향은 표본을 많이 얻을 수 없는 연구에서 더욱 뚜렸하다. 본 연구에서는 모평균과 모비율에 대한 검정에서 제1종 오류뿐만 아닌 제2종 오류를 고려한 경우 필요한 표본 크기를 결정하는 과정을 살펴보고 이를 웹사이트를 통해 공개하였다. 또한 주어진 표본 크기와 유의수준에 의한 검정력 계산도 함께 공개하였다.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2005년도 Proceedings of ISRS 2005
• /
• pp.26-29
• /
• 2005

In order to compute cloud coverage statistics over Asian region, an operational scheme for masking cloud-contaminated pixels in Advanced Very High Resolution Radiometer (AVHRR) daytime data was developed, evaluated and presented. Dynamic thresholding was used with channell, 2 and 3 to automatically create a cloud mask for a single image. Then the IO-day cloud coverage imagery was generated over the whole Asian region along with cloud-free composite imagery. Finally the monthly based statistics were computed based on the derived cloud coverage imagery in terms of land cover and country. As a result, it was found that 20-day is required to acquire the cloud free data over the whole Asia using NOAA AVHRR. The to-day cloud coverage and cloud-free composite imagery derived in this research is available via the web-site http://webpanda.iis.u-tokyo.ac.jp/CloudCover/.

• Communications for Statistical Applications and Methods
• /
• 제28권3호
• /
• pp.217-232
• /
• 2021

The Korea film industry has matured and the number of movie-watching per capita has reached the highest level in the world. Since then, movie industry growth rate is decreasing and even the total sales of movies per year slightly decreased in 2018. The number of moviegoers is the first factor of sales in movie industry and also an important factor influencing additional sales. Thus it is important to predict the number of movie audiences. In this study, we predict the cumulative number of audiences of films using stacking, an ensemble method. Stacking is a kind of ensemble method that combines all the algorithms used in the prediction. We use box office data from Korea Film Council and web comment data from Daum Movie (www.movie.daum.net). This paper describes the process of collecting and preprocessing of explanatory variables and explains regression models used in stacking. Final stacking model outperforms in the prediction of test set in terms of RMSE.

• 정보교육학회논문지
• /
• 제11권1호
• /
• pp.11-19
• /
• 2007

7차 교육 과정은 학교 운영에 학생, 교사, 학부모와 지역 사회의 요구를 적극 수렴하고 이를 반영할 것을 요구하고 있다. 이에 본 연구는 학교 운영을 효율적으로 하기 위한 웹 설문 시스템을 설계하고 구현하고자 한다. 본 연구에서 제시한 웹기반 설문 시스템은 관리자(조사자)와 응답자로 구분하여 설계하고 관리자는 인터넷만 연결되면 설문을 관리하고 결과를 확인할 수 있도록 하였다. 또한 문항별로 관련된 자료 데이터베이스를 제공하여 별도의 프로그램없이 본 시스템 내에서 실시간으로 설문 작성과 통계처리가 되도록 설계하였다. 본 시스템은 학교의 홈페이지에 한 메뉴로 활용하거나 단독 시스템으로 운영이 가능하여 학교 단위의 설문 조사를 쉽고 빠르게 할 수 있도록 해 준다.

• 부모자녀건강학회지
• /
• 제5권1호
• /
• pp.38-49
• /
• 2002

The purpose of this study was to investigate the needs of a mother when getting the information from the child care web sites. The subjects of this study were collected through on-line and off-line and constituted of 157 mothers who had neonates and infancy. All the mothers could use the internet. On-line subjects were 135 mothers who were the members of "Child Rearing" circle of internet portal site, Daum (www.daum.net) and off-line subjects were 22 mothers who visited the pediatric outpatient department. The instrument had 15 items about needs for information of the child care web sites. 12 items were multiple choice-questions and 3 items were open questions. In data analysis, SPSSWIN 10.0 program was utilized for descriptive statistics. The results were as follows. 1. The most useful web content that mothers represented was food and nutrition(24.4%) and the other rated as following order; growth and development(21.1%), supports among parents(16.8%). 2. The most great web problem that mothers represented was depth of content(43 people) and the other rated as following order; objectivity, specialty and reliability of content(29 people), no problems (23 people), too much advertisements(17 people), obligation on member(4 people). 3. Mothers represented the highest information needs about play and education(65 people) and the other rated as following order; nutrition(44 people), managements of common health problems(39 people), experiences of other parents(10 people), child caring methods (38 people), exchanges of things(7 people). 4. Mothers represented the highest consultation needs about common health problem (62 people) and the other rated as following order; growth and activity(60 people), play and education(30 people), nutrition(22 people), child caring methods(9 people).

• 한국정보통신학회논문지
• /
• 제18권2호
• /
• pp.252-259
• /
• 2014

2009년부터 공공기관, 교육기관, 의료기관, 복지시설의 국내 웹 접근성 의무화를 시작하여 2015년까지 문화예술, 법인까지 단계적으로 의무화 적용을 계획하고 있다. 본 연구에서는 국내 언론사(방송사, 종합일간지, 스포츠신문, 인터넷신문) 사이트의 웹 접근성 평가를 통하여, 국내 언론사들이 어느 정도 웹 사이트 접근성을 준수하고 있는지를 KWA-H 4.3 자동평가도구에서 제공하는 6개 자동점검 지침으로 평가하고, 통계적인 측면에서 분석해 보았다. 비모수적 분석인 크루스칼-왈리스(Kruskal-Wallis) 검정 결과, 기본언어 명시 항목의 경우 방송사> 인터넷신문> 종합일간지> 스포츠신문 순으로 높은 준수율(p값=0.041)을 보였으며, 레이블 제공 항목에 대해서는 스포츠신문> 방송사> 종합일간지> 인터넷신문 순으로 준수율(p값=0.015)이 나타났다.

• Communications for Statistical Applications and Methods
• /
• 제7권2호
• /
• pp.457-463
• /
• 2000

Recently, statistical analysis tools have been changed to the applications on the World Wide Web that access data stored in databases. On-line analytical processing(OLAP) is a class of technologies that give users statistical information with multidimensional views of data in databases. In this paper, we introduce the concept and requisites of OLAP system, and we propose some research issues.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권3호
• /
• pp.962-975
• /
• 2020

The accurate segmentation of infant brain MR image into white matter (WM), gray matter (GM), and cerebrospinal fluid (CSF) is very important for early studying of brain growing patterns and morphological changes in neurodevelopmental disorders. Because of inherent myelination and maturation process, the WM and GM of babies (between 6 and 9 months of age) exhibit similar intensity levels in both T1-weighted (T1w) and T2-weighted (T2w) MR images in the isointense phase, which makes brain tissue segmentation very difficult. We propose a deep network architecture based on U-Net, called Triple Residual Multiscale Fully Convolutional Network (TRMFCN), whose structure exists three gates of input and inserts two blocks: residual multiscale block and concatenate block. We solved some difficulties and completed the segmentation task with the model. Our model outperforms the U-Net and some cutting-edge deep networks based on U-Net in evaluation of WM, GM and CSF. The data set we used for training and testing comes from iSeg-2017 challenge (http://iseg2017.web.unc.edu).

• Communications for Statistical Applications and Methods
• /
• 제16권6호
• /
• pp.979-987
• /
• 2009

In this study, we compare the assessments made by self, peer and instructor in a statistics course. The goal is to investigate the following two questions: (1) Is it reasonable or fair to expect students to be responsible for assessing the work of their colleagues and themselves? (2) What are students' opinions about the learning effect after they participate in the assessment process? As part of the study investigating these questions, we designed a prototype for a Web-based assessment tool and a procedure to apply the assessment techniques in a statistics course. In addition, we collected and analyzed the data produced in the assessment processes from students and the instructor. The analysis results are summarized as follows: First, self assessment was not accord with instructor assessment, but peer assessment was similar to the assessment by instructor. This result reflected that it is reasonable or fair to expect students to be responsible for assessing the work of their colleagues. Second, peer assessment of their colleagues successfully helped students increase their understanding of the course, and the students increased their skills in the actual assessment process by assessing the work of their colleagues. Finally, many students indicated a high interest level on the assessments.

• 한국정보과학회논문지:컴퓨팅의 실제 및 레터
• /
• 제16권4호
• /
• pp.405-414
• /
• 2010

2008년도 SecruityFocus 자료에 따르면 마이크로소프트사의 인터넷 익스플로러를 통한 클라이언트 측 공격(client-side attack)이 50%이상 증가하였다. 본 논문에서는 가상머신 환경에서 능동적으로 웹 페이지를 방문하여 행위 기반(즉, 상태변경 기반)으로 악성 URL을 분석하여 탐지하고, 블랙리스트 기반으로 악성 URL을 필터링하는 시스템을 구현하였다. 이를 위해, 우선 크롤링 시스템을 구축하여 대상 URL을 효율적으로 수집하였다. 특정 서버에서 구동되는 악성 URL 탐지 시스템은, 수집한 웹페이지를 직접 방문하여 머신의 상태 변경을 관찰 분석하고 악성 여부를 판단한 후, 악성 URL에 대한 블랙리스트를 생성 관리한다. 웹 클라이언트 머신에서 구동되는 악성 URL 필터링 시스템은 블랙리스트 기반으로 악성 URL을 필터링한다. 또한, URL의 분석 시에 메시지 박스를 자동으로 처리함으로써, 성능을 향상시켰다. 실험 결과, 게임 사이트가 다른 사이트에 비해 악성비율이 약 3배 많았으며, 파일생성 및 레지스트리 키 변경 공격이 많음을 확인할 수 있었다.