• East Asian mathematical journal
• /
• 제5권1호
• /
• pp.95-110
• /
• 1989

In a general variance component model, nonnegative quadratic estimators of the components of variance are considered which are invariant with respect to mean value translaion and have minimum bias (analogously to estimation theory of mean value parameters). Here the minimum is taken over an appropriate cone of positive semidefinite matrices, after having made a reduction by invariance. Among these estimators, which always exist the one of minimum norm is characterized. This characterization is achieved by systems of necessary and sufficient condition, and by a cone restricted pseudoinverse. In models where the decomposing covariance matrices span a commutative quadratic subspace, a representation of the considered estimator is derived that requires merely to solve an ordinary convex quadratic optimization problem. As an example, we present the two way nested classification random model. An unbiased estimator is derived for the mean squared error of any unbiased or biased estimator that is expressible as a linear combination of independent sums of squares. Further, it is shown that, for the classical balanced variance component models, this estimator is the best invariant unbiased estimator, for the variance of the ANOVA estimator and for the mean squared error of the nonnegative minimum biased estimator. As an example, the balanced two way nested classification model with ramdom effects if considered.

• 대한간호학회지
• /
• 제45권3호
• /
• pp.429-438
• /
• 2015

Purpose: This study was conducted to develop key performance indicators (KPIs) for home care nursing (HCN) based on a balanced scorecard, and to construct a performance prediction model of strategic objectives using the Bayesian Belief Network (BBN). Methods: This methodological study included four steps: establishment of KPIs, performance prediction modeling, development of a performance prediction model using BBN, and simulation of a suggested nursing management strategy. An HCN expert group and a staff group participated. The content validity index was analyzed using STATA 13.0, and BBN was analyzed using HUGIN 8.0. Results: We generated a list of KPIs composed of 4 perspectives, 10 strategic objectives, and 31 KPIs. In the validity test of the performance prediction model, the factor with the greatest variance for increasing profit was maximum cost reduction of HCN services. The factor with the smallest variance for increasing profit was a minimum image improvement for HCN. During sensitivity analysis, the probability of the expert group did not affect the sensitivity. Furthermore, simulation of a 10% image improvement predicted the most effective way to increase profit. Conclusion: KPIs of HCN can estimate financial and non-financial performance. The performance prediction model for HCN will be useful to improve performance.

• 제어로봇시스템학회논문지
• /
• 제5권6호
• /
• pp.667-675
• /
• 1999

In this paper, we propose a robust controller design of RTP(Rapid Thermal Processing) system using structured uncertainty approach. Using the weighted mixed sensitivity function, we solve the robust stability problem against disturbance and temperature variation, and design a $\mu$ controller using curve fitting method against structured uncertainty. Also the reduction method should be requried because of the difficulty of implementaion with the obtained high order controller. We dal with robust stability and performance of RTP system by the design of $\mu$ controller for original model and Schur balanced reduced model. Finally the simulation results are proposed to show the validity of the proposed method.

• 한국병원경영학회지
• /
• 제21권4호
• /
• pp.1-12
• /
• 2016

The purpose of this study are suggest to performance measurement model of Electronic Medical Record(EMR) and Key Performance Index(KPI). For data collection, 665 questionnaires were distributed to medical record administrators and insurance reviewers at 31 hospitals, and 580 questionnaires were collected(collection rate: 87.2%). Regarding methodology, Critical Success Factor(CSF) and index of the information system were derived based on previous studies, and these were set as performance measurement factors of EMR system. The performance measurement factors were constructed by perspective using BSC, and analysis on causal relationship between factors was conducted. A model of causal relationship was established, and performance measurement model of EMR system was proposed through model validation. Analysis on causal relationship between performance management factors revealed that utility cognition of the learning & growth perspective factor had causal relationship with job efficiency(${\beta}=0.20$) and decision support(${\beta}=0.66$) of the internal process perspective factors, and security had causal relationship with system satisfaction(${\beta}=0.31$) of the customer perspective factor. System quality had causal relationship with job efficiency(${\beta}=0.66$) and decision support(${\beta}=0.76$) of the internal process perspective factors, all of which were statistically significant(P<0.01). Job efficiency of the internal process perspective had causal relationship with system satisfaction(${\beta}=0.43$), and decision support had causal relationship with decision support satisfaction(${\beta}=0.91$) and job satisfaction (${\beta}=0.74$), all of which were statistically significant(P<0.01). System satisfaction of the customer perspective had causal relationship with job satisfaction(${\beta}=0.12$), job satisfaction had causal relationship with cost reduction(${\beta}=0.53$) of the financial perspective, and decision support satisfaction had causal relationship with productivity improvement(${\beta}=0.40$)of the financial perspective(P<0.01). Also, cost reduction of the financial perspective had causal relationship with productivity improvement(${\beta}=0.37$), all which were statistically significant(P<0.05). Suitability index verification of the performance measurement model whose causal relationship was found to be statistically significant revealed that $X^2/df=2.875$, RMR=0.036, GFI=0.831, AGFI=0.810, CFI=0.887, NFI=0.838, IFI=0.888, RMSEA=0.057, PNFI=0.781, and PCFI=0.827, all of which were in suitable levels. In conclusion, the performance measurement indices of EMR system include utility cognition, security, and system quality of the learning & growth perspective, decision support and job efficiency of the internal process perspective, system satisfaction, decision support satisfaction, and job satisfaction of the customer perspective, and productivity improvement and cost reduction of the financial perspective. In this study, it is expected that the performance measurement indices and model of EMR system which are suggested by the author, will be a measurement tool available for system performance measurement of EMR system in medical institutions.

• 한국음향학회:학술대회논문집
• /
• 한국음향학회 1998년도 학술발표대회 논문집 제17권 2호
• /
• pp.35-38
• /
• 1998

입체음향 시스템의 방향감 제어에서 필수적인 머리전달함수(Head-Related Transfer Function)는 일반적으로 FIR 또는 IIR 필터로 구현되며, IIR 필터의 경우 FIR 보다 비교적 저차 모델링이 가능한 장점을 갖는다. 본 논문에서는 Balanced Model Reduction(BMR) 방법을 이용하여 비교적 높은 차수를 가지는 FIR 형태의 HRTF 필터를 IIR 필터로 설계하고, 입체음향 시스템의 실시간 구현시 필수적인 계산의 효율을 높이는 방법을 제시한다. 또한 IIR 필터로 근사화된 HRTF 필터의 주파수 패턴을 분석하여, 음상정위에 큰 영향을 미치는 주파수 대역을 좀 더 나은 해상도를 갖도록 설계하여 저차 모델 설계시 발생하는 오차가 음상정위에 미치는 영향을 최소화 하였다.

• Journal of Biosystems Engineering
• /
• 제1권1호
• /
• pp.1-6
• /
• 1976

This paper is the third one of the study on balanced type oscillating mole-drainer, the first one was presented in No 9. Gyeongsang College Report and the second one in Vol. 17, No.4 of the KSAE. In the first part of this study, the characteristics of traction forces between the nonoscillating earth working equipments and oscillating ones was compared. A model of the balanced type oscillating mole-drainer, which composed of a mechanism that may reduce the machine vibration, was designed following the dimensional analysis and similitude technique. The model test was carried out to clarify the balancing mechanism of the oscillating parts and other parts of the machine. In the light of the results from the model tests, a prototype machine was made for experimental purpose. Results from the field test by a reported in the near future. In the second report, the model tests were carried out under the same soil conditions, i.e, . oscillating frequency, running velocity, and oscillating amplitude, etc. It was clear that use of balanced type oscillating model could substantially reduce the vibration of the whole system of the machine, when compared with the nonoscillating type model. In this paper(the third report), results of investigation on the traction force, power requirement, and moment. etc, is presented. Analysis of variance technique was used for analyzing the effect of the frequency, amplitude, and running velocity on the draft force, torque, power requirements, and moments. The results obtained from the model tests are as follows, 1) By practicing a balanced-type oscillating mole-drainer, it was possible to reduce the traction resistance by 55.1-61. 2 percent of traction resistance, however, was 1.75 - 1.95 times greater than the value of resistance which was induced by use of a mole-drainer with single bullet. The resistance of rear shank against soil was considered as a main causing factor of the above results. 2) As the oscillation frequency was increased, the traction resistance was decreased. Considering on the effect of oscillation the greater the amplitude, and the slower the running velocity was, the greater the reduction ratio of traction resistance was. 3) The ratio of the traction resistance of oscillating mole-drainer to that of non-oscillating one could be represented as a function of dimensionless variable (V/$Af$). The results from the tests were well agreed with the reported results from the experim ents on oscillation plow or hoe. 4) By taking a lower value of (V/$Af$), reducing the traction resistance was possible. This fact meant, however, that the efficiency of mole drain practice would be lower. 5) It was experimentally confirmed under the same condition of soil that the variable (R/$rD1^3$) could be represented as a function of a variable($V^2/gD$) when a non\ulcornerocillating mole-drainer was used. 6) When a oscillating mole-drainer was used, the variable(R/$rD_1^{3}$) could be represented as a function of two variables ($v^2/gD_1$) and (V^2/gD_1$). 7) The torque was not affected by a change of frequency. However, a relation of proportionality existed between torque and amplitude, running velocity, and ratio of bullet diameter. When a balanced type oscillating mole-drainer with two bullets was used, torque was increased by 52.8-78. 4 percent and total power requirement was also increased. 8) Total power requirement was increased linearly in accordance with the increasing frequency, 41.96 percent of total power was used for oscillating action. The magnitude of total power requirement was 1. 8-9. 4 times greater than that of a non-oscillating mechanism. In the view point of power requirement, it was not advisable to increase the frequency, amplitude, running velocity, and ratio of bullet diameter at the same time. 9) Only the positive moment occured in the rear shank. Change of the diameter of a rear bullet, could not affect the balancing against the soil resistance. It was necessary for rear bullet to have a large resistance against soil only when the rear bullet was in backward motion. 10) Within an extent of the experimental base, optimum limits for several design factors were A=0.5cm, $f$=22.5Hz, V=O. 05m/sec, and $\lambda$=1.0 By adapting these values traction resistance was reduced by 40 percent and vibration acceleration wa s reduced by 60 percent. Even though the total , power requirements for operating a balanced type oscillation mechanism was greater ~than that of non-oscillating one, using a oscillating mechanism would be more effective. Because a balanced type oscillating mechanism is used, tractive resistance will be reduced and then the lighter . tractive equipment could be used.

• BMB Reports
• /
• 제42권9호
• /
• pp.617-622
• /
• 2009

We investigated the genetic associations of ischemic stroke by identifying epistasis of its heterogeneous subtypes such as small vessel occlusion (SVO) and large artery atherosclerosis (LAA). Epistasis was analyzed with 24 genes in 207 controls and 271 patients (SVO = 110, LAA = 95) using multifactor dimensionality reduction and entropy decomposition. The multifactor dimensionality reduction analysis with any of 1- to 4-locus models showed no significant association with LAA (P > 0.05). The analysis of SVO, however, revealed a significant association in the best 3-locus model with P10L of TGF-$\beta{1}$, C1013T of SPP1, and R485K of F5 (testing balanced accuracy = 63.17%, P < 0.05). Subsequent entropy analysis also revealed that such heterogeneity was present and quite a large entropy was estimated among the 3 loci for SVO (5.43%), but only a relatively small entropy was estimated for LAA (1.81%). This suggests that the synergistic epistasis model might contribute specifically to the pathogenetsis of SVO, which implies a different etiopathogenesis of the ischemic stroke subtypes.

• Journal of Multimedia Information System
• /
• 제6권4호
• /
• pp.225-234
• /
• 2019

Deep learning has become increasingly popular in both academic and industrial areas nowadays. Various domains including pattern recognition, Computer vision have witnessed the great power of deep neural networks. However, current studies on deep learning mainly focus on quality data sets with balanced class labels, while training on bad and imbalanced data set have been providing great challenges for classification tasks. We propose in this paper a method of data analysis-based data reduction techniques for selecting good and diversity data samples from a large dataset for a deep learning model. Furthermore, data sampling techniques could be applied to decrease the large size of raw data by retrieving its useful knowledge as representatives. Therefore, instead of dealing with large size of raw data, we can use some data reduction techniques to sample data without losing important information. We group PCB characters in classes and train deep learning on the ResNet56 v2 and SENet model in order to improve the classification performance of optical character recognition (OCR) character classifier.

• 전자공학회논문지SC
• /
• 제39권1호
• /
• pp.33-41
• /
• 2002

본 논문에서는 기존의 주파수하중 균형절단 기법과 주파수하중 한켈노옴 근사화 기법에 비하여 더 작은 H∞ 하중 축소오차를 가지는 새로운 알고리듬을 제시한다. 제시한 알고리듬은 제한 실 보조정리로부터 반복적인 두 단계의 선형행렬부등식 형태로 유도한다. 또한 제안한 알고리듬을 성능보장을 위한 제어기 차수축소기법에 적용한다. 수치적 예를 통하여 제안한 알고리듬의 타당성을 보이고 기존의 모델 차수축소기법과 비교 분석하며 HIMAT(highly maneuverable aircraft technology) 시스템의 예를 통하여 성능보장을 위한 제어기 차수축소 기법에 적용할 수 있음을 보인다.

• 한국콘텐츠학회논문지
• /
• 제5권6호
• /
• pp.74-86
• /
• 2005

본 논문에서는 2 채널 입체음향 시스템을 효율적으로 구현하기 위하여 고차의 FIR 필터로 구현된 머리 전달함수를 저차의 IIR 필터로 근사시키는 알고리즘을 제안한다. 제안하는 알고리즘은 균형화모델감소법의 개념을 바탕으로 한다. 근사화된 머리전달함수를 헤드폰을 통해 바이노럴 사운드로 재생하면 입체음향이 구현된다. 더미 헤더의 머리전달함수를 512차 FIR 필터에서 32차 IIR 필터로 근사화하여 기존의 머리전달함수와 성능을 비교한다. 이를 위해 10명을 대상으로 음상 정위에 대한 실험을 한다. 실험은 컴퓨터 시뮬레이션과 TMS320C32를 이용한 하드웨어 실험을 병행한다. 실험을 통해 기존의 FIR 필터를 사용한 경우와 같이 IIR 필터로 근사화된 머리전달함수를 사용한 경우도 동일하게 인지되는 것을 확인할 수 있다.