• Journal of the Korean Statistical Society
• /
• 제27권4호
• /
• pp.407-420
• /
• 1998

This paper is concerned with selecting covariates to be included in building linear random effects models designed to analyze clustered response normal data. It is based on a Bayesian approach, intended to propose and develop a procedure that uses probabilistic considerations for selecting premising subsets of covariates. The approach reformulates the linear random effects model in a hierarchical normal and point mass mixture model by introducing a set of latent variables that will be used to identify subset choices. The hierarchical model is flexible to easily accommodate sign constraints in the number of regression coefficients. Utilizing Gibbs sampler, the appropriate posterior probability of each subset of covariates is obtained. Thus, In this procedure, the most promising subset of covariates can be identified as that with highest posterior probability. The procedure is illustrated through a simulation study.

• 품질경영학회지
• /
• 제26권1호
• /
• pp.143-160
• /
• 1998

This article is concerned with the selection of subsets of predictor variables to be included in bulding the binary response logistic regression model. It is based on a Bayesian aproach, intended to propose and develop a procedure that uses probabilistic considerations for selecting promising subsets. This procedure reformulates the logistic regression setup in a hierarchical normal mixture model by introducing a set of hyperparameters that will be used to identify subset choices. It is done by use of the fact that cdf of logistic distribution is a, pp.oximately equivalent to that of $t_{(8)}$/.634 distribution. The a, pp.opriate posterior probability of each subset of predictor variables is obtained by the Gibbs sampler, which samples indirectly from the multinomial posterior distribution on the set of possible subset choices. Thus, in this procedure, the most promising subset of predictors can be identified as that with highest posterior probability. To highlight the merit of this procedure a couple of illustrative numerical examples are given.

• 대한의용생체공학회:의공학회지
• /
• 제44권3호
• /
• pp.176-181
• /
• 2023

During cancer treatment, the patient's response to drugs appears differently at the cellular level. In this paper, an image-based cell phenotypic feature quantification and key feature selection method are presented to predict the response of patient-derived cancer cells to a specific drug. In order to analyze the viability characteristics of cancer cells, high-definition microscope images in which cell nuclei are fluorescently stained are used, and individual-level cell analysis is performed. To this end, first, image stitching is performed for analysis of the same environment in units of the well plates, and uneven brightness due to the effects of illumination is adjusted based on the histogram. In order to automatically segment only the cell nucleus region, which is the region of interest, from the improved image, a superpixel-based segmentation technique is applied using the fluorescence expression level and morphological information. After extracting 242 types of features from the image through the segmented cell region information, only the features related to cell viability are selected through the ReliefF algorithm. The proposed method can be applied to cell image-based phenotypic screening to determine a patient's response to a drug.

• 한국IT서비스학회지
• /
• 제6권2호
• /
• pp.125-143
• /
• 2007

IT Services market is growing rapidly in the business industry and SOA-based Web Services have been introduced as an effective vehicle for the integration of enterprise-wide applications within organizations. The number of publicly available Web Services is ever increasing recently in a variety of areas, and as the number of public Web Services increases, there will be many Web Services with the same functionality. These services, however, will vary in their QoS properties, such as price, response time and availability, and it is very important to choose a right service while satisfying given QoS constraints. This paper addresses the issue of selecting composite Web Services which involves conditional branches in business processes. It is essential to have any conditional branches satisfy the global QoS constraints at service selection phase, since the branches are chosen to execute at run-time dynamically. We proposed service selection procedures for basic structure of conditional branches and explained them by examples. Experiments were conducted to analyze the impact of the number of candidate services and service types on the time of finding service solutions.

• Communications for Statistical Applications and Methods
• /
• 제27권5호
• /
• pp.535-546
• /
• 2020

In genetic association studies, pleiotropy is a phenomenon where a variant or a genetic region affects multiple traits or diseases. There have been many studies identifying cross-phenotype genetic associations. But, most of statistical approaches for detection of pleiotropy are based on individual tests where a single variant association with multiple traits is tested one at a time. These approaches fail to account for relations among correlated variants. Recently, multivariate regularization methods have been proposed to detect pleiotropy in analysis of high-dimensional genomic data. However, they suffer a problem of tuning parameter selection, which often results in either too many false positives or too small true positives. In this article, we applied selection probability to multivariate regularization methods in order to identify pleiotropic variants associated with multiple phenotypes. Selection probability was applied to individual elastic-net, unified elastic-net and multi-response elastic-net regularization methods. In simulation studies, selection performance of three multivariate regularization methods was evaluated when the total number of phenotypes, the number of phenotypes associated with a variant, and correlations among phenotypes are different. We also applied the regularization methods to a wild bean dataset consisting of 169,028 variants and 17 phenotypes.

• 한국컴퓨터정보학회논문지
• /
• 제19권9호
• /
• pp.125-139
• /
• 2014

본 논문에서는 준지도 지지 벡터 회귀 모델(semi-supervised support vector regression)을 이용한 반응 모델링(response modeling)을 제안한다. 반응 모델링의 성능 및 수익성을 높이기 위해, 고객 데이터 셋의 대부분을 차지하는 레이블이 존재하지 않는 데이터를 기존 레이블이 존재하는 데이터와 함께 학습에 이용한다. 제안하는 알고리즘은 학습 복잡도를 낮은 수준으로 유지하기 위해 일괄 학습(batch learning) 방식을 사용한다. 레이블 없는 데이터의 레이블 추정에서 불확실성(uncertainty)을 고려하기 위해, 분포추정(distribution estimation)을 하여 레이블이 존재할 수 있는 영역을 정의한다. 그리고 추정된 레이블 영역으로부터 오버샘플링(oversampling)을 통해 각 레이블이 없는 데이터에 대한 레이블을 복수 개 추출하여 학습 데이터 셋을 구성한다. 이 때, 불확실성의 정도에 따라 샘플링 비율을 다르게 함으로써, 불확실한 영역에 대해 더 많은 정보를 발생시킨다. 마지막으로 지능적 학습 데이터 선택 기법을 적용하여 학습 복잡도를 최종적으로 감소시킨다. 제안된 반응 모델링의 성능 평가를 위해, 실제 마케팅 데이터 셋에 대해 다양한 레이블 데이터 비율로 실험을 진행하였다. 실험 결과 제안된 준지도 지지 벡터 회귀 모델을 이용한 반응 모델이 기존 모델에 비해 더 높은 정확도 및 수익을 가질 수 있다는 점을 확인하였다.

• Journal of the Korean Statistical Society
• /
• 제30권4호
• /
• pp.585-595
• /
• 2001

A second-order response surface model is often used to approximate the relationship between a response factor and a set of explanatory factors. In this article, we deal with canonical analysis in response surface models. For the interpretation of the geometry of second-order response surface model, standard errors and confidence intervals for the eigenvalues of the second-order coefficient matrix play an important role. If the confidence interval for some eigenvalue includes 0 or the estimate of some eigenvalue is very small (near to 0) with respect to other eigenvalues, then we are able to delete the corresponding canonical factor. We propose a formulation of criterion which can be used to select canonical factors. This criterion is based on the IMSE(=Integrated Mean Squared Error). As a result of this method, we may approximately write the canonical factors as a set of some important explanatory factors.

• Journal of Ecology and Environment
• /
• 제36권3호
• /
• pp.151-158
• /
• 2013

We estimated phenotypic selection on the height and number of flowering-stalks in a rosette annual Cardamine hirsuta by applying path analysis to the data collected at three natural populations located in central Japan. The path from rosette size was positively connected with the fruit production through the both height and number of flowering-stalks. In the all three populations, the paths from the number of stalks were more strongly connected with the fruit production than from the height of stalks. The paths from the rosette size showed similar magnitude with the number of stalks and the height of stalks. The direct path from rosette size to the fruit production was detected only at one site. These results suggest stronger phenotypic selection on the rosette size through the number of stalks than the height of stalks. The lateral branching rather than increment of individual inflorescence size is the major response to control the fruit production for C. hirsuta growing in a natural habitat.

• International Journal of Naval Architecture and Ocean Engineering
• /
• 제5권2호
• /
• pp.227-245
• /
• 2013

Reliable strength assessment of the Liquefied Natural Gas (LNG) cargo containment system under the sloshing impact load is very difficult task due to the complexity of the physics involved in, both in terms of the hydrodynamics and structural mechanics. Out of all those complexities, the proper selection of the design sloshing load which is applied to the structural model of the LNG cargo containment system, is one of the most challenging one due to its inherent randomness as well as the statistical analysis which is tightly linked to the design sloshing load selection. In this study, the response based strength assessment procedure of LNG cargo containment system has been developed and proposed as an alternative design methodology. Sloshing pressure time history, measured from the model test, is decomposed into wavelet basis function targeting the minimization of the number of the basis function together with the maximization of the numerical efficiency. Then the response of the structure is obtained using the finite element method under each wavelet basis function of different scale. Finally, the response of the structure under entire sloshing impact time history is rapidly calculated by synthesizing the structural response under wavelet basis function. Through this analysis, more realistic response of the system under sloshing impact pressure can be obtained without missing the details of pressure time history such as rising pattern, oscillation due to air entrapment and decay pattern and so on. The strength assessment of the cargo containment system is then performed based on the statistical analysis of the stress peaks selected out of the obtained stress time history.

• International Journal of Contents
• /
• 제5권1호
• /
• pp.51-55
• /
• 2009

In this paper, we present a structure of personalized e-Learning system to study for a test formalized by uniform multiple-choice using on/off line mixed estimations as is the case of Driver :s License Test in Korea. Using the system a candidate can study toward the license through the Internet (and/or mobile instruments) within the personalized concept based on IRT(item response theory). The system accurately estimates user's ability parameter and dynamically offers optimal evaluation problems and learning contents according to the estimated ability so that the user can take possession of the license in shorter time. In order to establish the personalized e-Learning concepts, we build up 3 databases and 2 agents in this system. Content DB maintains learning contents for studying toward the license as the shape of objects separated by concept-unit. Item-bank DB manages items with their parameters such as difficulties, discriminations, and guessing factors, which are firmly related to the learning contents in Content DB through the concept of object parameters. User profile DB maintains users' status information, item responses, and ability parameters. With these DB formations, Interface agent processes user ID, password, status information, and various queries generated by learners. In addition, it hooks up user's item response with Selection & Feedback agent. On the other hand, Selection & Feedback agent offers problems and content objects according to the corresponding user's ability parameter, and re-estimates the ability parameter to activate dynamic personalized learning situation and so forth.