• 대한화학회지
• /
• 제58권6호
• /
• pp.543-552
• /
• 2014

A quantitative structure activity relationship (QSAR) study is performed for modeling and prediction of oral bioavailabilities of 216 diverse set of drugs. After calculation and screening of molecular descriptors, linear and nonlinear models were developed by using multiple linear regression (MLR), artificial neural network (ANN), support vector machine (SVM) and random forest (RF) techniques. Comparison between statistical parameters of these models indicates the suitability of SVM over other models. The root mean square errors of SVM model were 5.933 and 4.934 for training and test sets, respectively. Robustness and reliability of the developed SVM model was evaluated by performing of leave many out cross validation test, which produces the statistic of $Q^2_{SVM}=0.603$ and SPRESS = 7.902. Moreover, the chemical applicability domains of model were determined via leverage approach. The results of this study revealed the applicability of QSAR approach by using SVM in prediction of oral bioavailability of drugs.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2001년도 추계학술대회논문집 I
• /
• pp.252-256
• /
• 2001

Scattering of surface materials has been known as one of the most important aspects in evaluating the acoustics of concert halls are designed. One of the methods that can reduce the errors in estimating the reverberation time and other acoustic parameters through computer modeling is to calculate scattering coefficient of surface materials. However. so far, no objective and reliable methods measuring scattering coefficient has been suggested. In this situation, ISO has suggested the method of measuring the random-incidence scattering coefficient on surfaces in diffuse field, whereas AES has introduced a method on directional-incidence in free field. In this study, the scattering coefficients of five kinds of hemispheres (1.5, 2.0. 2.5. 3.0. 3.5cm) were measured by using the ISO method in 1：10 reverberation chamber. It was found that 3.0cm hemisphere has the highest scattering coefficient satisfying 95％ reliability.

• 한국안전학회지
• /
• 제15권1호
• /
• pp.167-175
• /
• 2000

Work performance and human error are complicated phenomena so that it is very difficult to grasp the true nature of them. However, Event Related Potential (ERP) may give a clue to them because human brain reflects diverse psychophysiological process. In the present study, the possibility of ERP application to the ergonomic area was evaluated in view of grasping error symptoms. For that purpose, the subjects were asked to count specific characters in a random character matrix on a computer monitor, and their ERP was compared with their performance data. Based upon the results, the amplitude of P300 was not so high as that in the case of the Odd Ball tasks, correct response corresponded with stable ERP with high P300 amplitude whereas wrong response did with unstable, fluctuating ERP with low P300 amplitude. Those results coincided with the work performance, and it was concluded that 3-wave fluorescent with illumination level of 800 lux would be recommendable for the counting task in concern. Conclusively, ERP including P300 might supply an objective clue to the problem of human errors in cognitive process.

• Smart Structures and Systems
• /
• 제19권4호
• /
• pp.383-391
• /
• 2017

In this paper, we investigate the impacts of network topology on the performance of a distributed estimation algorithm, namely combine-then-adaptive (CTA) diffusion LMS, based on the data with or without the assumptions of temporal and spatial independence with noisy links. The study covers different network models, including the regular, small-world, random and scale-free whose the performance is analyzed according to the mean stability, mean-square errors, communication cost (link density) and robustness. Simulation results show that the noisy links do not cause divergence in the networks. Also, among the networks, the scale free network (heterogeneous) has the best performance in the steady state of the mean square deviation (MSD) while the regular is the worst case. The robustness of the networks against the issues like node failure and noisier node conditions is discussed as well as providing some guidelines on the design of a network in real condition such that the qualities of estimations are optimized.

• 대한기계학회논문집B
• /
• 제26권12호
• /
• pp.1707-1714
• /
• 2002

This paper deals with the development of a new method that can obtain heat transfer coefficient and reference free stream temperature simultaneously, The method is based on transient heat transfer experiments using two narrow-band TLCs. The method is validated through error analysis in terms of the random uncertainties in the measured temperatures. It is found that the errors could be reduced more than 2 times less. The general method described in this paper is applicable to many heat transfer models with unknown free stream temperature.

• 대한기계학회논문집A
• /
• 제21권2호
• /
• pp.327-336
• /
• 1997

This paper addresses an analytical approach to the mechanical error analysis and tolerance design of a four-bar path generator with lubricated joints. The mobility method is applied to consider lubrication effects and the four-bar path generator is stochastically modeled by using the clearance vector model for methanical error analysis. To show the validity of the proposed method, the mechanical errors obtained by applying the method to a four-bar path generator are compared with those by Monte Carlo simulation. Based on this analytical method, an optimal tolerance design problem is formulated and solved for the four-bar path generator.

• Communications for Statistical Applications and Methods
• /
• 제3권1호
• /
• pp.73-85
• /
• 1996

In this paper, we consider the distribution-free confidence intervals for the reliability of the stress-strength model when the stress X and strength Y depend linearly on some explanatory variables z and w, respectively. We apply these confidence intervals to the Rocket-Motor data and compare the results to those of Guttman et al. (1988). Some simulation results show that the distribution-free confidence intervals have better performance for nonnormal errors compared to those of Guttman et al. (1988), which are designed for normal random variables in respect that the former yield the coverage levels closer to the nominal coverage level than the latter.

• Communications for Statistical Applications and Methods
• /
• 제21권2호
• /
• pp.193-200
• /
• 2014

It was recognized by some researchers that the disturbance variance in a censored regression model is frequently underestimated by the maximum likelihood method. This underestimation has implications for the estimation of marginal effects and asymptotic standard errors. For instance, the actual coverage probability of the confidence interval based on a maximum likelihood estimate can be significantly smaller than the nominal confidence level; consequently, a Bayesian estimation is considered to overcome this difficulty. The behaviors of the maximum likelihood and Bayesian estimators of disturbance variance are examined in a fixed effects panel regression model with a limited dependent variable, which is known to have the incidental parameter problem. Behavior under random effect assumption is also investigated.

• 한국가시화정보학회지
• /
• 제4권2호
• /
• pp.59-68
• /
• 2006

We have evaluated the performances of the following six interpolation schemes used for win-dow deformation in particle image velocimetry (PIV): the linear, quadratic, B-spline, cubic, sinc, Lagrange interpolations. Artificially generated images comprised of particles of diameter in a range pixel were investigated. Three particle diameters were selected for detailed evaluation: pixel with a constant particle concentration $0.02particle/pixel^2$. Two flow patterns were considered: uniform and shear flows. The mean and random errors, and the computation times of the interpolation schemes were determined and compared.

• 소음진동
• /
• 제2권2호
• /
• pp.99-106
• /
• 1992

The complex Young's modulus of a viscoelastic material can be obtained as a function of frequency from the measurements of relative motion between the two ends of a bar-type specimen. Non-resonance method is usually used to obtain the complex Young's modulus over wide range of frequency including resonance points, while in resonance method information at resonance frequencies only is used. However, the complex Young's modulus obtained by the non-resonance method is often unreliable in the anti-resonance frequency regions because of the measurement noise problems. In this study, the effects of the random measurement errors on estimating the complex Young's modulus are studied in the aspect of sensitivity, and how to obtain the reliable frequency region for a given measurement error level is shown. The usable frequency regions in determining the complex Young's modulus are represented by a non-dimensional parameter formed with the wave length and specimen length.