• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2003년도 ICCAS
• /
• pp.1521-1524
• /
• 2003

In this paper, we propose a "Distance Correlation Selection operator (DCS)" as a new selection operator. For Genetic Algorithm (GA), many improvements have been proposed. The MGG (Minimal Generation Gap) model proposed by Satoh et.al. shows good performance. The MGG model has all advantages of conventional models and the ability of avoiding the premature convergence and suppressing the evolutionary stagnation. The proposed method is an extension of selection operator in the original MGG model. Generally, GA has two types of selection operators, one is "selection for reproduction", and the other is "selection for survival"; the former is for crossover and the latter is the individuals which survive to the next generation. The proposed method is an extension of the former. The proposed method utilizes distance information between individuals. From this extension, the proposed method aims to expand a search area and improve ability to search solution. The performance of the proposed method is examined with several standard test functions. The experimental results show good performance better than the original MGG model.

• Journal of the Korean Data and Information Science Society
• /
• 제20권6호
• /
• pp.1191-1197
• /
• 2009

Most Bayesian approaches to model selection of wavelet analysis have drawbacks that computational cost is expensive to obtain accuracy for the fitted unknown function. To overcome the drawback, this article introduces loss functions which are criteria for level dependent threshold selection in wavelet based Bayesian methods with arbitrary size and regular design points. We demonstrate the utility of these criteria by four test functions and real data.

• 기술혁신연구
• /
• 제2권1호
• /
• pp.116-141
• /
• 1994

This paper presents a model for R&D project selection and evaluation system design developed through literature review. The model emphasizes the fitness between the five elements of the system : evaluation phase and purpose, personnel and organization, evaluation critiria and decision model, evaluation form and procedure, and projects. The model was applied in real situation as a test case. The important findings are that a good project selection and evaluation model contributes only partially to the effectiveness of the project selection and that system development and implementation activity is a dynamic and multi-facetted learning process.

• 응용통계연구
• /
• 제34권6호
• /
• pp.863-874
• /
• 2021

FLSA는 총변동벌점을 이용해 구간별상수인 평균 구조를 구현하는 벌점모형으로 다중변화점 탐색을 위해 활용되고 있다. 한편, FLSA는 변화점 탐색에 있어서 점근적 일치성이 만족되지 않으므로 잡음의 크기가 0에 가깝게 수렴하는 경우에도 다수의 거짓 변화점이 식별될 수 있다는 단점이 있다. 이 연구에서는 이러한 FLSA의 문제점을 해결하기 위한 사후추론 방법으로 순열검정 방법을 제안한다. 단일변화점 모형과 관련된 순열검정 방법은 Antoch와 Hušková (2001)에 의해 제안된 바 있다. 이 연구에서는 Antoch와 Hušková (2001)의 검정절차를 확장하여 다중변화점 식별에 사용되는 FLSA와 결합함으로써 다중변화점 모형에 적용할 수 있는 순열검정절차를 제안한다. 모의실험 결과, 제안된 방법은 z-검정과 CUSUM 통계량의 극한분포에 기반을 둔 검정방법에 비해 전반적으로 우수하였으며 거짓 변화점의 식별에 유용함을 확인할 수 있었다.

• Genomics & Informatics
• /
• 제14권4호
• /
• pp.187-195
• /
• 2016

In genetic association studies with high-dimensional genomic data, multiple group testing procedures are often required in order to identify disease/trait-related genes or genetic regions, where multiple genetic sites or variants are located within the same gene or genetic region. However, statistical testing procedures based on an individual test suffer from multiple testing issues such as the control of family-wise error rate and dependent tests. Moreover, detecting only a few of genes associated with a phenotype outcome among tens of thousands of genes is of main interest in genetic association studies. In this reason regularization procedures, where a phenotype outcome regresses on all genomic markers and then regression coefficients are estimated based on a penalized likelihood, have been considered as a good alternative approach to analysis of high-dimensional genomic data. But, selection performance of regularization procedures has been rarely compared with that of statistical group testing procedures. In this article, we performed extensive simulation studies where commonly used group testing procedures such as principal component analysis, Hotelling's $T^2$ test, and permutation test are compared with group lasso (least absolute selection and shrinkage operator) in terms of true positive selection. Also, we applied all methods considered in simulation studies to identify genes associated with ovarian cancer from over 20,000 genetic sites generated from Illumina Infinium HumanMethylation27K Beadchip. We found a big discrepancy of selected genes between multiple group testing procedures and group lasso.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2003년도 춘계학술대회 논문집
• /
• pp.5-8
• /
• 2003

This paper presents a methodology to be able to select an appropriate RP system that suits the end use of a part. Evaluation factors used in process selection include major attributes such as accuracy, roughness, strength, elongation, part cost and build time that greatly affect the performance of RP systems. Crisp values such as accuracy and surface roughness are obtained with a new test part developed. The test part is designed with conjoint analysis to reflect users' preference. The part cost and build time that have approximate ranges due to cost and many variable parameters are presented by linguistic values that can be described with triangular fuzzy numbers. Based on the evaluation values obtained, an appropriate RP process for a specific part application is selected by using the modified TOPSIS(Technique of Order Preference by Similarity to Ideal Solution) method. It uses crisp data as well as linguistic variables, and each weight on the alternatives is assigned by using pair-wise comparison matrix. The ranking order helps the decision making of the selection of RP systems.

• 전기학회논문지
• /
• 제57권6호
• /
• pp.1058-1062
• /
• 2008

In clinical decision support system(CDSS), unlike rule-based expert method, appropriate data-driven machine learning method can easily provide the information of individual feature(clinical test) for disease classification. However, currently developed methods focus on the improvement of the classification accuracy for diagnosis. With the analysis of feature importance in classification, one may infer the novel clinical test sets which highly differentiate the specific diseases or disease states. In this background, we introduce a novel CDSS that integrate a classifier and feature selection module together. Random forest algorithm is applied for the classifier and the feature importance measure. The system selects the significant clinical tests discriminating the diseases by examining the classification error during backward elimination of the features. The superior performance of random forest algorithm in clinical classification was assessed against artificial neural network and decision tree algorithm by using breast cancer, diabetes and heart disease data in UCI Machine Learning Repository. The test with the same data sets shows that the proposed system can successfully select the significant clinical test set for each disease.

• 한국항공운항학회지
• /
• 제28권3호
• /
• pp.74-83
• /
• 2020

Securing excellent pilots is not only directly linked to the military's improved combat capabilities, but also a way to minimize human and property losses from aircraft accidents. Therefore, a scientific method is needed to diagnose pilot aptitude from the pilot selection process and select those with high accident potential, those who are dropped out of the flight training process, and those who are not suitable for pilot life in advance. Developed countries have implemented pilot aptitude tests to solve these problems early on, but so far, the Korean Army has not introduced a pilot aptitude test system that uses diagnostic tools in the helicopter pilot selection process. Therefore, in this study, scientific diagnostic tools are developed for selecting helicopter pilots, and through this, it is predicted that the number of people who are likely to be dropped out of the training course and who have the potential for accidents will be selected in advance and eliminated in the selection process. In this context, prior research examined the key factors involved in the pilot aptitude test. Through this, the aptitude test items were developed and aptitude tests were conducted on student pilots currently in flight training, and the results of flight training were analyzed.

• 동아시아경상학회지
• /
• 제8권4호
• /
• pp.1-18
• /
• 2020

Purpose - The current study attempts to reveal the causal relationship and identify the interrelationships among individual value, product selection attribute, and re-purchase intention of consumers when purchasing kimchi. Research design, data, and methodology - The surveys were distributed and retrieved to kimchi consumers in Seoul and Gyeonggi Province who were over 20 years old and who had purchased kimchi before. The current study utilized the self-reporting survey as the research method. Result - First, as a result of hypothesis 1 test, both internal value and external value had a significant effect. Second, as a result of hypothesis 2 test, while internal value had a significant effect, external value did not have a significant effect. Third, as a result of hypothesis 3 test, while internal value had a significant effect, external value did not have a significant effect. Fourth, as a result of hypothesis 4 test, all quality, brand, and price had a significant effect. Fifth, as a result of hypothesis 5 test, while internal value had a significant effect, external value did not have a significant effect. Conclusion - The current study aims to establish marketing and service strategies to maximize profits and secure competitive advantage in the kimchi industry through changes in the management strategy of the Korean kimchi industry.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2005년도 추계종합학술대회
• /
• pp.901-904
• /
• 2005

Testing of mixed-signal circuit has become a difficult task for test engineers and efficient test solution to these problems are needed. In this paper a new specification-based mixed-signal test method called TSS(Time Slot Specification) using high performance current sensors and a novel test point selection technique without heavy computational overhead are proposed. External output and power nodes are used for test points and accessed by the current sensors in the ATE.