• 응용통계연구
• /
• 제37권2호
• /
• pp.225-237
• /
• 2024

경쟁위험자료에서 일부 공변량들이 연구대상들의 일부분에 대해 관측되지 않을 수 있다. 그런 경우 결측된 공변량 값을 가진 연구대상들을 분석에서 제외하는 것은 편향된 추정치와 효율성 손실이 발생할 수 있다. 본 논문에서는 누락된 공변량을 가진 원인별 비례위험모형의 회귀모수 추정을 위해 다중대체 방법과 증대된 역 확률 가중 방법을 연구하였다. 모의실험을 통해 다중대체 방법과 증대된 역 확률 가중 방법에 의해 구해진 추정량의 성능을 평가한 결과, 이 방법들이 잘 수행됨을 확인하였다. 미국 국립암연구소의 전립선, 폐, 대장, 난소 암 선별 시험 연구에서 제공하는 종양 크기의 값이 누락된 유방암 자료에 대해 암 사망 위험률과 다른 원인 사망 위험률에 유의한 영향을 미치는 요인을 파악하기 위해 다중대체 방법과 증대된 역 확률 가중 방법을 적용하였다. 다중대체 방법과 증대된 역 확률 가중 방법에 의해 원인별 비례위험모형을 적합한 결과, 인종, 기혼여부, 병기, 분화도, 종양의 크기는 유방암 사망 위험률에 유의한 영향을 미치는 요인들이였으며, 병기가 유방암 사망 위험률을 높이는데 가장 큰 영향을 미치는 요인임을 확인하였다. 진단시 연령과 종양의 크기는 다른 원인 사망 위험률을 높이는데 유의한 영향을 미치는 요인이였다.

• 대한기계학회논문집A
• /
• 제21권8호
• /
• pp.1195-1208
• /
• 1997

Bellows is a component in piping systems which absorbs mechanical deformation with flexibility. Its geometry is an axial symmetric shell which consists of two toroidal shells and one annular plate or conical shell. In order to analyze bellows, this study presents the finite element analysis using a conical frustum shell element. A finite element analysis is developed to analyze various bellows. The validity of the developed program is verified by the experimental results for axial and lateral stiffness. The formula for calculating the natural frequency of bellows is made by the simple beam theory. The formula for fatigue life is also derived by experiments. The shape optimal design problem is formulated using multiple objective optimization. The multiple objective functions are transformed to a scalar function by weighting factors. The stiffness, strength and specified stiffness are considered as the multiple objective function. The formulation has inequality constraints imposed on the fatigue limit, the natural frequencies, and the manufacturing conditions. Geometric parameters of bellows are the design variables. The recursive quadratic programming algorithm is selected to solve the problem. The results are compared to existing bellows, and the characteristics of bellows is investigated through optimal design process. The optimized shape of bellows is expected to give quite a good guideline to practical design.

• 한국지능시스템학회논문지
• /
• 제20권1호
• /
• pp.101-108
• /
• 2010

본 논문에서는 퍼지 k-NN과 reconstruction error에 기반을 둔 feature selection을 이용한 lazy 분류기 설계를 제안하였다. Reconstruction error는 locally linear reconstruction의 평가 지수이다. 새로운 입력이 주어지면, 퍼지 k-NN은 local 분류기가 유효한 로컬 영역을 정의하고, 로컬 영역 안에 포함된 데이터 패턴에 하중 값을 할당한다. 로컬 영역과 하중 값을 정의한 우에, feature space의 차원을 감소시키기 위하여 feature selection이 수행된다. Reconstruction error 관점에서 우수한 성능을 가진 여러 개의 feature들이 선택 되어 지면, 다항식의 일종인 분류기가 하중 최소자승법에 의해 결정된다. 실험 결과는 기존의 분류기인 standard neural networks, support vector machine, linear discriminant analysis, and C4.5 trees와 비교 결과를 보인다.

• 한국지반공학회:학술대회논문집
• /
• 한국지반공학회 2000년도 가을 학술발표회 논문집
• /
• pp.665-672
• /
• 2000

The problems of discontinuous layer interface are very important in the algorithm and programming for the analysis of multi-layered consolidation using a numerical analysis, finite difference method(F.D,M.). Better results can be obtained by the process for discontinuous layer interface, since it can help consolidation analysis to model the actual ground Explicit method is simple for analysis algorithm and convenient for use except for applying the operator Crank-Nicolson method represents implicit method, which have different analysis method according to weighting factor. This method uses different algorithm according to dimension. And, this paper uses alternative direction implicit method. The purpose of this paper provides an efficient computer algorithm based on numerical analysis using finite difference method which account for multi-layered soils with confined aquifer to determine the degree of consolidation and excess pore pressures relative to time and positions more realistically.

• 한국경영과학회지
• /
• 제28권4호
• /
• pp.39-60
• /
• 2003

In this study, we evaluate financial performance of 21 domestic life insurers using SAW (simple additive weighting), ELECTREII, cluster analysis respectively, and suggest a hybrid approach of combining cluster analysis and ELECTREII to reclassify the life insurers into more meaningful groups according to their respective financial features. We also perform the sensitivity analysis employing ANOVA and Tukey's test to examine the robustness of ELECTREII, which would be influenced by decision maker's subjective preference parameters. Consequently, it is shown that ELECTREII turns out to be a flexible method providing decision makers with useful ranking Information especially under fuzzy decision making situation with incomparable alternatives, and hence it can serve as a complementary method to overcome the weakness of classical cluster analysis.

• 한국도로학회논문집
• /
• 제15권1호
• /
• pp.135-141
• /
• 2013

PURPOSES: In this study, as part of an effort to develop HUD for public transit, it is proposed that the decision of order priority of contents which will be disposed to bus drivers through HUD for public transit using AHP(Analytic Hierarchy Process) technique. METHODS: In AHP analysis method brainstorming, factor analysis, hierarchical structuring, and weighting analysis were performed by applying a classical analysis method. RESULTS: By the result of analysis it is shown that unlike car drivers, bus drivers prefer information related to bus intervals, bus stop, and door open and close to information related to vehicle running. Also, bus stop information and bus interval information were ranked as first and second place in order priority of HUD contents for public transit by experts. CONCLUSIONS: This method of selecting order priority of HUD contents for public transit can provide a basic foundation for selecting order priority of traffic information contents as well as other HUD contents.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1994년도 Proceedings of the Korea Automatic Control Conference, 9th (KACC) ; Taejeon, Korea; 17-20 Oct. 1994
• /
• pp.82-87
• /
• 1994

In this study, dynamic optimal design far a two degree-of-freedom anthropomorphic robot module is performed. Several dynamic design indices associated with the inertia matrix and the inertia power array are introduced. Analysis for the relationship between the dynamic parameters and the design indices shows that trade-offs exist between the isotropy and the dynamic design indices related to the actuator size. A composite design index is employed to deal with multi-criteria based design with different weighting factors, in a systematic manner. We demonstrate the fact that dynamic optimization is another significant step to enhance the system performances, followed by kinematic optimization.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2002년도 ICCAS
• /
• pp.79.3-79
• /
• 2002

$\textbullet$ Introduction $\textbullet$ Structure of variable multi-valued spatial filter detector(VMSFD) $\textbullet$ Moment analysis by VMSFD $\textbullet$ Experimental results $\textbullet$ Fabricated chip with photodiode array $\textbullet$ Conclusion

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 1997년도 춘계학술대회논문집; 경주코오롱호텔; 22-23 May 1997
• /
• pp.493-499
• /
• 1997

In this study, the changes in transmitting and receiving characteristics, especially resonance frequencies, of the underwater acoustic planar array transducer by the transformers used in the analog weighting circuit are investigated. Electrical equivalent circuit analysis shows that an ideal transformer does not change the resonance frequency of the transmitting mode, but the resonance frequency which gives the maximum receiving sensitivity can be designed by adjusting the magnitude of reactance of transformer.

• 한국음향학회지
• /
• 제3권1호
• /
• pp.25-35
• /
• 1984

This paper analysis the Dynamic Time Warping algorithms for time normalization of speech pattern and discusses the Dynamic Programming algorithm for spoken Korean isolated digits recognition. In the DP matching, feature vectors of the reference and test pattern are consisted of first three formant frequencies extracted by power spectrum density estimation algorithm of the ARMA model. The major differences in the various DTW algorithms include the global path constrains, the local continuity constraints on the path, and the distance weighting/normalization used to give the overall minimum distance. The performance criterias to evaluate these DP algorithms are memory requirement, speed of implementation, and recognition accuracy.