• Genomics & Informatics
• /
• 제17권4호
• /
• pp.41.1-41.12
• /
• 2019

Survival analysis mainly deals with the time to event, including death, onset of disease, and bankruptcy. The common characteristic of survival analysis is that it contains "censored" data, in which the time to event cannot be completely observed, but instead represents the lower bound of the time to event. Only the occurrence of either time to event or censoring time is observed. Many traditional statistical methods have been effectively used for analyzing survival data with censored observations. However, with the development of high-throughput technologies for producing "omics" data, more advanced statistical methods, such as regularization, should be required to construct the predictive survival model with high-dimensional genomic data. Furthermore, machine learning approaches have been adapted for survival analysis, to fit nonlinear and complex interaction effects between predictors, and achieve more accurate prediction of individual survival probability. Presently, since most clinicians and medical researchers can easily assess statistical programs for analyzing survival data, a review article is helpful for understanding statistical methods used in survival analysis. We review traditional survival methods and regularization methods, with various penalty functions, for the analysis of high-dimensional genomics, and describe machine learning techniques that have been adapted to survival analysis.

• Nuclear Engineering and Technology
• /
• 제52권1호
• /
• pp.211-221
• /
• 2020

One fundamental element of probabilistic safety assessment (PSA) is the initiating event (IE) analysis. Since IE frequencies can change over time, time-trend analysis is required to obtain optimized IE frequencies. Accordingly, such time-trend analyses have been employed to estimate industry-average IE frequencies for use in the PSAs of U.S. nuclear power plants (NPPs); existing PSAs of Korean NPPs, however, neglect such analysis in the estimation of IE frequencies. This article therefore provides the method for and results of estimating Korean industry-average IE frequencies using time-trend analysis. It also examines the effects of the IE frequencies obtained from this study on risk insights by applying them to recently updated internal events Level 1 PSA models (at-power and shutdown) for an OPR-1000 plant. As a result, at-power core damage frequency decreased while shutdown core damage frequency increased, with the related contributions from each IE category changing accordingly. These results imply that the incorporation of time-trend analysis leads to different IE frequencies and resulting risk insights. The IE frequency distributions presented in this study can be used in future PSA updates for Korean NPPs, and should be further updated themselves by adding more recent data.

• 한국해양공학회지
• /
• 제20권5호
• /
• pp.36-41
• /
• 2006

The main object of this study is to develop an accurate and convenient method for the response analysis of offshore structures in real sea states. A numerical procedure is described for predicting the motion responses and tension variations of the ISSC TLP in multi-directional irregular waves. The developed numerical approach in the frequency domain is based on acombination of the three dimensional source distribution method, the dynamic response analysis method, and the spectral analysis method. Frequency domain analysis in the multi-directional irregular waves is expanded to a time domain analysis by using a convolution integral after obtaining the impulse response by Fourier transformation. The results of the comparison between responses in the frequency and time domain confirmed the validity of the proposed approach.

• 한국측량학회지
• /
• 제34권6호
• /
• pp.569-578
• /
• 2016

With the recent advancement of surveying and technology, the spatial data acquisition rates and precision have been improved continually. As the updates of spatial data are rapid, and the size of data increases in line with the advancing technology, the LOD (Level of Detail) algorithm has been adopted to process data expressions in real time in a streaming format with spatial data divided precisely into separate steps. The existing GRID analysis utilizes the single DEM, as it is, in examining and analyzing all data outside the analysis area as well, which results in extending the analysis time in proportion to the quantity of data. Hence, this study suggests a method to reduce analysis time and data throughput by acquiring and analyzing DEM data necessary for GRID analysis in real time based on the area of analysis and the level of precision, specifically for streaming DEM data, which is utilized mostly for 3D geographic information service.

• 가족자원경영과 정책
• /
• 제9권2호
• /
• pp.41-57
• /
• 2005

The purpose of this study was to explore various types of couples' time use and to investigate the characteristics of each type. The subject of the study consisted of 96 employed wives and their husbands and 215 unemployed wives and their husbands living in Seoul. The data were analyzed by descriptive statistics and cluster analysis using SAS 8.2 program. The major findings were as follows: The time use of wives and their husbands both on one weekday and one holiday was classified as personal time, paid work time, housework time, and free time. These 4 categories of time were basic elements to classify the types of couples' time use in the cluster analysis. As a result of the cluster analysis, four types of time use of couples were yielded. These were called as 'personal time oriented type', 'paid work time oriented type', 'housework time oriented type' and 'free time oriented type'.

• 대한전자공학회논문지TC
• /
• 제41권11호
• /
• pp.47-55
• /
• 2004

본 논문에서는 이동 통신망에서의 향상된 시간 기반의 위치 갱신 방법을 제안하고 제안된 방법의 성능을 분석하였다. 분석에서는 포아송 분포의 호 도착시간을 가정하였고, 지수분포의 셀 체류시간을 가정하였으며 분석결과로서는 제안된 방법에서의 최적의 갱신주기를 구하였다. 부가적으로 갱신주기의 값의 변화에 따른 비용의 특성을 관찰하고자 갱신주기의 값이 0 또는 무한대로 근접할 경우에 대하여 위치관리 비용을 분석하였다. 제안된 위치갱신방법과 기존에 제안된 오리지날 시간 기반의 위치갱신방법과의 비교를 위하여 오리지날 시간 기반의 위치갱신 방법에 대한 분석도 수행하였다. 두 방법에 대한 비교에서 제안된 방법이 주어진 환경에서 성능이 향상됨을 알 수 있었다.

• Tribology and Lubricants
• /
• 제39권2호
• /
• pp.35-42
• /
• 2023

Bearing is a mechanical component that supports loads and transmits rotation. As the application of high-value-added products such as semiconductors, aviation, and robots have recently become diverse and more precise, an accurate bearing performance prediction and evaluation technology is required. Bearing performance evaluation can be divided into evaluations based on bearing theory and on numerical analysis. An evaluation based on numerical analysis is a technique that has been highlighted because the problems that remained unsolved owing to time problems can be solved through recent developments in computers. However, current studies have the disadvantage of not considering the essential changes over time and bearing rotation. In this study, bearing performance evaluation based on rigid body dynamic analysis considering rotation and load over time is performed. Rigid body dynamic analysis is performed for deep groove ball bearing to calculate the load applied by the ball. The reliability of the analysis is verified by comparing it with the results calculated using bearing theory. In addition, rigid body dynamic analysis is performed for automotive wheel bearings to calculate the contact angle and load applied by the ball for cases where axial load and radial load are applied, respectively. The effect of rotation and load over time is evaluated from these results.

• 한국환경과학회지
• /
• 제18권4호
• /
• pp.369-374
• /
• 2009

To examine the trend on the flowering time in some weather flora including Prunus serrulata var. spontanea, Cosmos bipinnatus, and Robinia pseudo-acacia in Busan, the changes in time series and rate of flowering time of plants were analyzed using the method of time series analysis. According to the correlation between the flowering time and the temperature, changing pattern of flowering time was very similar to the pattern of the temperature, and change rate was gradually risen up as time goes on. Especially, the change rate of flowering time in C. bipinnatus was 0.487 day/year and showed the highest value. In flowering date in 2007, the difference was one day between measurement value and prediction value in C. bipinnatus and R. pseudo-acacia, whereas the difference was 8 days in P. mume showing great difference compared to other plants. Flowering time was highly related with temperature of February and March in the weather flora except for P. mume, R. pseudo-acacia and C. bipinnatus. In most plants, flowering time was highly related with a daily average temperature. However, the correlation between flowering time and a daily minimum temperature was the highest in Rhododendron mucronulatum and P. persica, otherwise the correlation between flowering time and a daily maximum temperature was the highest in Pyrus sp.

• 가정과삶의질연구
• /
• 제22권1호
• /
• pp.139-151
• /
• 2004

The purposes of this study were to examine the time use on weekdays and weekends by the time amount of each activities and compare time amounts of single parent with that of two parents family. So two methods were applied, one was secondary data analysis ,the other was deep-interview. The major findings of analysis of time use were as follows; 1) Mothers in single parent family used significantly less time amount of household labor and family care than them in two parent family 2) Fathers in single parent family used significantly more time amount of household labor but used significantly less time amounts of family care than them in two parent family And 3) Mothers in single parent family used significantly more time amounts of household labor and family care than fathers in single parent family. The results of deep-interview were as follows; roughly the amounts of household labor time, family care time, leisure time were of little quantity and absence of spouse, family size, family life cycle stage, economic condition, sex role identity, psychological factor maybe seemed important.

• 한국멀티미디어학회논문지
• /
• 제9권5호
• /
• pp.671-680
• /
• 2006

'최소 순회 시간’ 방법은 ‘시간 넷’(Timed Net)에서 모든 변천(transition)들을 최소한 한번 이상 격발하고 초기 마킹으로 되돌아오는데 걸리는 최소 시간을 찾아주는 수학적 방법으로, 컴퓨터 시스템 분석에 널리 사용된다. 시간 넷은 변천에 지연 시간이 연합된 변형된 페트리 넷으로 실세계의 사건이 실행되는데 걸리는 시간이 일반적으로 확률적 시간인데 반하여 시간 넷에 사용되는 지연시간은 상수이기 때문에 최소 순회 시간 방법에 의한 분석 결과는 현실성이 부족하다. 그래서 본 논문은 사건의 실행 시간을 확률적 숫자로 표현하는 ‘확률적 시간 넷'을 제안하고 ‘확률적 시간 넷'의 최소 순회 시간 분석 방법을 소개한다. 또한 ‘확률적 시간 넷’의 적용 예로 위치 기반 서비스를 제공하는 멀티미디어 시스템을 ‘확률적 시간 넷’으로 모델링하고 분석한다. 기존의 최소 순회 시간 분석 결과의 전형적인 형태는 '최소 10 단위 시간 이상 걸린다'와 같은 형태인데 반하여 ‘확률적 시간 넷’의 분석 결과의 형태는 '10 단위 시간 이내일 확률이 85% 이다.'와 같은 확률적 형태이다.