• The Journal of the Acoustical Society of Korea
• /
• v.31 no.2
• /
• pp.87-92
• /
• 2012

In this paper, a robust feature compensation method to deal with the environmental mismatch is proposed. The proposed method applies energy based weights according to the degree of speech presence to the Mean subtraction, Variance normalization, and ARMA filtering (MVA) processing. The weights are further smoothed by the moving average and maximum filters. The proposed feature compensation algorithm is evaluated on AURORA 2 task and distant talking experiment using the robot platform, and we obtain error rate reduction of 14.4 % and 44.9 % by using the proposed algorithm comparing with MVA processing on AURORA 2 task and distant talking experiment, respectively.

• Journal of Nutrition and Health
• /
• v.27 no.6
• /
• pp.583-590
• /
• 1994

The longitudinal changes in the transitional milk yields of Korean lactating women(14 primiparae, 11 multiparae) have been studied by test-weighing method in the part of Kangwon Province form 7 day to 15 days postpartum. The transitional milk yields at 7, 10, 15 days postpartum were 531$\pm$148,598$\pm$156 and 639$\pm$169g, respectively. The mean milk yield was 589$\pm$162g from 7 to 15 days postpartum. The transitional milk yields between primiparae and multiparae appeared not significantly different but significantly different between mothers of boys and girl(p<0.05). The distribution of individual transitional milk yields were found 550-649g(28.0%), 450-549g(24.0%), 650-749g(13.3%) and 750-849g(13.3%). The transitional milk yields were not affected by mother's age, weight gain during pregnancy, gestational period and infant's weight at birth but affected by maternal height(p<0.05). Maternal weight reduction during the lactation had no correlation with the transitional milk yeilds.

• Communications for Statistical Applications and Methods
• /
• v.13 no.1
• /
• pp.191-204
• /
• 2006

This paper develops a Bayesian method to derive the optimal sequential preventive maintenance(PM) policy by determining the PM schedules which minimize the mean cost rate. Such PM schedules are derived based on a general sequential imperfect PM model proposed by Lin, Zuo and Yam(2000) and may have unequal length of PM intervals. To apply the Bayesian approach in this problem, we assume that the failure times follow a Weibull distribution and consider some appropriate prior distributions for the scale and shape parameters of the Weibull model. The solution is proved to be finite and unique under some mild conditions. Numerical examples for the proposed optimal sequential PM policy are presented for illustrative purposes.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2013.04a
• /
• pp.353-358
• /
• 2013

Present paper deals with turbulence-airfoil interaction noise and mainly investigates the effects of airfoil thickness on the broadband noise spectrum. The acoustic power radiation from an airfoil is predicted using high-order time-domain method, which is based on the computational aeroacoustic technique solving the linear Euler equations. The homogeneous and isotropic turbulence is generated by utilizing the synthetic turbulence modeling based on random particle method. The airfoils taken into consideration are a flat-plate and a NACA0012 airfoil aligned with uniform mean flow. The effects of airfoil thickness on the radiated inflow turbulence noise are investigated by comparing acoustic power spectrum predicted for each airfoil. The comparison of acoustic power spectrum reveals that the airfoil thickness significantly contributes the high frequency noise reduction.

• Journal of Powder Materials
• /
• v.1 no.2
• /
• pp.135-144
• /
• 1994

Silicon carbide powder was prepared from mixtures of Sangdong silica sand and carbon black by SHS (Self propagating High temperature Synthesis) method which utilizes magnesiothermic reduction of silica. In the powder preparation process, the reacted powder was leached by chloric acid to remove the magnesium oxide and was subsequently roasted to remove free carbon. The impurities were mostly eliminated by hot acid treatment. The resultant SiC powder showed the mean particle size of 0.22 ${\mu}{\textrm}{m}$ and the specific surface area of $66.55 m^2/g$. The SiC powder was mixed with 1 wt% of boron and of carbon to increase densification rate. The mixed powder was pressed and sintered pressurelessly at $2100^{\circ}C$ for 30 min in argon gas. The sintered body showed the hardness of $2550 kg{\cdot}f/mm^2$ and the fracture toughness, KIC of $3.47 MN/m^{3/2}$.

• Journal of the Korea Society for Simulation
• /
• v.9 no.4
• /
• pp.67-75
• /
• 2000

Often system analysts are interested in the estimation of percentile for system performance. For instance, in PERT network system, the percentile that the project. Typically the control variate method is used to reduce the variability of mean response using the correlation between the response and the control variates with a little additional cost during the course of simulation. In the same spirit, we apply this method to estimate the percentile of project completion time in PERT system, and evaluate the efficiency of the controlled estimator for its percentile.1 Simulation results indicate that the controlled estimators are more effective in reducing the variances of estimators than the simple estimators, however those tend to a little underestimate the percentiles for some critical values. We need more simulation experiments to examine such a kind of bias problem. We expect this research presents a step forward in the area of variance reduction techniques of stochastic simulation.

• Journal of Information Processing Systems
• /
• v.14 no.4
• /
• pp.904-915
• /
• 2018

We demonstrate how social media content can be used to predict the unemployment rate, a real-world indicator. We present a novel method for predicting the unemployment rate using social media analysis based on natural language processing and statistical modeling. The system collects social media contents including news articles, blogs, and tweets written in Korean, and then extracts data for modeling using part-of-speech tagging and sentiment analysis techniques. The autoregressive integrated moving average with exogenous variables (ARIMAX) and autoregressive with exogenous variables (ARX) models for unemployment rate prediction are fit using the analyzed data. The proposed method quantifies the social moods expressed in social media contents, whereas the existing methods simply present social tendencies. Our model derived a 27.9% improvement in error reduction compared to a Google Index-based model in the mean absolute percentage error metric.

• Journal of the Korean Society for Precision Engineering
• /
• v.16 no.3 s.96
• /
• pp.215-221
• /
• 1999

The object of the thermal error compensation system in machine tools is improving the accuracy of a machine tool through real time error compensation. The accuracy of the machine tool totally depends on the accuracy of thermal error model. A thermal error model can be obtained by appropriate combination of temperature variables. The proposed method for optimal variable selection in the thermal error model is based on correlation grouping and successive regression analysis. Collinearity matter is improved with the correlation grouping and the judgment function which minimizes residual mean square is used. The linear model is more robust against measurement noises than an engineering judgement model that includes the higher order terms of variables. The proposed method is more effective for the applications in real time error compensation because of the reduction in computational time, sufficient model accuracy, and the robustness.

• The korean journal of orthodontics
• /
• v.21 no.1 s.33
• /
• pp.17-29
• /
• 1991

The purpose of this study was to provide quantitative data describing the effect of rapid palatal expansion (RPE) on nasal airway resistance (NAR). RPE is an orthopedic procedure which is commonly used to widen the maxilla in skeletal Class III patients. 18 subjects (9 males and 9 females, mean age: 10 years 7 months) were selected from the Orthodontics in Yongdong Severance Hospital. Recordings of NAR were taken by active anterior method prior to expansion, immediately after desired maximum expansion, and after approximately 3 months and 6 months, and 1 year. All data was recorded and statistically processed with the SPSS program of IBM PC system. The results are as followings . 1. The average initial NAR of the subjects was 3.84 cm $H_2O/LPS\;(SD{\pm}1.34)$. It was greater than the average NAR of the normal subjects. 2. Among 18 subjects, 9 subjects showed reduction of NAR and 9 subjects showed no specific change of NAR after expansion. 3 An average reduction in NAR after immediately expansion was statistically significant at the 0.05 level. 4. The reduction appeared stable throughout the post treatment observation period of 1 year after expansion. From these results, it was suggested that RPE is a useful method to reduce NAR.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2011.05a
• /
• pp.70-70
• /
• 2011

기후변화 연구는 불확실한 미래를 전망하는 과정이므로 '불확실성'은 모든 기후변화 영향평가의 키워드임에 분명하다. 하지만 불확실성 평가를 위해 IPCC에서 제공되고 있는 수많은 GCM 시나리오를 모두 활용하기에는 많은 시간과 노력이 필요하기 때문에 이를 효율적으로 수행할 수 있는 방법이 필요하다 본 연구에서는 시나리오 저감(scenario reduction)방법을 이용하여, 수많은 GCM 시나리오 대신 몇 개의 대표적 GCM 시나리오로도 충분히 불확실성을 유지할 수 있는 시나리오 저감(scenario reduction)방법을 수립하고 제시하였다. IPCC 기후시나리오 중 20C3M과 A & B 배출시나리오를 바탕으로 생산되는 71개의 GCM 시나리오를 다운로드 받아 월평균 기온과 강수량에 대하여 한반도를 대상으로 분석하였다. 비교결과, 기온 전망은 실측과 비슷한 경향성을 보였으나 강수량은 홍수기를 모의하지 못하는 것으로 나타났다. 시나리오 저감방법은 시나리오 분류(scenario cluster)방법과 시나리오 선정(scenario selection) 방법으로 구성된다. 시나리오 분류방법에서는 k-mean방법을 이용하여 5개의 cluster로 나누었으며, 시나리오 선정방법에서는 GCM 시나리오 선정기법을 조사 분석하여 연구방향과 목적에 따라 GCM 시나리오 선정기법을 선택할 수 있는 표를 제시하고, 이 중 시나리오의 확률밀도함수를 이용하는 PDF method를 적용하였다. 본 연구에서는 불확실성 정량화를 위해 maximum entropy를 이용하였다. 또한 시나리오 저감방법이 불확실성을 유지하는지 비교하기 위해 PDF method를 이용하여 정확성이 높은 순으로 5개의 GCM 시나리오를 선정(best 시나리오)하여 불확실성을 정량화하였다. GCM 시나리오의 분산을 이용하여 maximum entropy를 산정한 결과, 20C3M 배출시나리오에서는 모든 시나리오의 entropy는 3.08, 시나리오 저감방법은 2.75, best 시나리오는 2.28이었으며, 이는 시나리오 저감방법은 모든 시나리오의 89.3%의 불확실성을 설명하고 있으나 best 시나리오는 74.0%밖에 설명하지 못한다는 것을 나타낸다. A & B 배출시나리오에서도 시나리오 저감 방법을 사용한 GCM 시나리오가 best 시나리오보다 모든 시나리오의 불확실성을 더 잘 설명하는 것으로 나타났다. 이와 같이 수많은 GCM 시나리오를 사용하는 것보다 몇 개의 대표 시나리오를 이용하여 기후 변화 불확실성을 유지하면서 미래전망을 할 수 있다면, 매우 효율적으로 기후변화 연구를 수행할 수 있을 것으로 사료된다.