• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.45 no.5
• /
• pp.13-20
• /
• 2008

Although digital hearing aid algorithms have been developed to compensate hearing loss and to help hearing impaired people to communicate with others, digital hearing aid user still complain about difficulty of hearing the speech. The reason could be the quality of speech through digital hearing aid is insufficient to understand the speech caused by feedback, residual noise and etc. And another thing is masking effect among formants that makes sound quality low. In this study, we measured the masking characteristics of normal listeners and hearing impaired listeners having presbyacusis to confirm masking effect in speech itself. The experiment is composed of 5 tests; pure tone test, speech reception threshold (SRT) test, word recognition score (WRS) test, puretone masking test and speech masking test. In speech masking test, there are 25 speeches in each speech set. And log likelihood ratio (LLR) is introduced to evaluate the distortion of each speech objectively. As a result, the speech perception became lower by increasing the quantity of formant enhancement. And each enhanced speech in a speech set has statistically similar LLR, however speech perception is not. It means that acoustic masking effect rather than distortion influences speech perception. In actuality, according to the result of frequency analysis of the speech that people can not answer correctly, level difference between first formant and second formant is about 35dB, and it is similar to result of pure tone masking test(normal hearing subject:36.36dB, hearing impaired subject:32.86dB). Characteristics of masking effect is not similar between normal listeners and hearing impaired listeners. So it is required to check the characteristics of masking effect before wearing a hearing aid and to apply this characteristics to fitting.

• The Journal of the Acoustical Society of Korea
• /
• v.10 no.5
• /
• pp.69-76
• /
• 1991

본 논문에서는 통계적 방법에 의한 음소의 자동분할에 관한 알고리즘을 제안하였다. 우선 음성 신호를 AR 모델로 모델링한 후 스펙트럼이 변화하기 전과 변화한 후의 모델에 대해서 likelihood ratio 와 mutual information을 고려한 test statistics 로부터 모델 계수가 변화하는 곳을 예측해 내고 이 곳을 음소의 경계로 판단한다. 이 경우 검파되지 못하는 대부분의 음소는 짧은 자음이었으며 Signed front-to-back maximum area ratio을 이용하여 개선하였다. 또한 false alarm error을 줄이기 위해 두 segment 사이의 distortion 으로부터 smoothing을 하였다. 3명의 화자에 대한 실험 결과 non-detection error는 10%, false alarm error는 20% 정도로 나타났지만 화자간에 알고리즘의 성능 변화가 거의 없으 며 특히 분할된 경계치 분포는 전체 음소의 90% 이상이 이 30ms 이내에 위치하였다.

• Journal of Korean Academy of Nursing
• /
• v.43 no.2
• /
• pp.154-164
• /
• 2013

Purpose: The purpose of this article is twofold: 1) introducing logistic regression (LR), a multivariable method for modeling the relationship between multiple independent variables and a categorical dependent variable, and 2) examining use and reporting of LR in the nursing literature. Methods: Text books on LR and research articles employing LR as main statistical analysis were reviewed. Twenty-three articles published between 2010 and 2011 in the Journal of Korean Academy of Nursing were analyzed for proper use and reporting of LR models. Results: Logistic regression from basic concepts such as odds, odds ratio, logit transformation and logistic curve, assumption, fitting, reporting and interpreting to cautions were presented. Substantial shortcomings were found in both use of LR and reporting of results. For many studies, sample size was not sufficiently large to call into question the accuracy of the regression model. Additionally, only one study reported validation analysis. Conclusion: Nursing researchers need to pay greater attention to guidelines concerning the use and reporting of LR models.

• Asian-Australasian Journal of Animal Sciences
• /
• v.15 no.3
• /
• pp.315-320
• /
• 2002

Test-day records (n=13677) sampled from 896 ewes in 5-9 (${\mu}$=7.5) monthly test-days were used to estimate genetic and phenotypic parameters of test-day yields, lactation milk yield (TMY), length of the milking period (DAYS) and three measures of persistency of milk yield in Boutsico dairy sheep. Τhe measures of persistency were the slope of the regression line (${\beta}$), the coefficient of variation (CV) of the test-day milk yields and the maximum to average daily milk yield ratio (MA). The estimates of variance components were obtained under a linear mixed model by restricted maximum likelihood. The heritability of test-day yields ranged from 0.15 to 0.24. DAYS were found to be heritable ($h^2$=0.11). Heritability estimates of ${\beta}$, CV and MA were 0.15, 0.13, 0.10, respectively. Selection for maximum lactation yields is expected to result in prolonged milking periods, high rates of decline of yields after peak production, variable test-day yields and higher litter sizes. Selection for flatter lactation curves would reduce lactation yields, increase slightly the length of the milking period and decrease yield variation as well as litter size. The most accurate prediction of TMY was obtained with a linear regression model with the first five test-day records.

• Asian-Australasian Journal of Animal Sciences
• /
• v.24 no.10
• /
• pp.1335-1340
• /
• 2011

Different genetic parameters for weaning weight and wool traits were estimated using restricted maximum likelihood (REML) in Angora rabbits. Total wool yield of first (I), second (II) and third (III) clips were taken as a separate trait under study. The records from more than 2,700 animals were analysed through fitting six animal models with various combinations of direct and maternal effects. A log likelihood ratio test was used to select the most appropriate model for each trait. Direct heritability estimates for the wool traits were found to be moderate to high across different models. Heritability estimates obtained from the best model were 0.24, 0.22, 0.20 and 0.21 for weaning weight, clip I, II and III; respectively. Maternal effects especially due to permanent environment had higher importance at clip I and found to be declining in subsequent clips. The estimates of repeatability of doe effect on wool traits were 0.44, 0.26 and 0.18 for clip I, II and III; respectively. Weaning weight had moderately high genetic correlations with clip I (0.57) and II (0.45), but very low (0.11) with clip III. Results indicated that genetic improvement for wool yield in Angora rabbit is possible through direct selection. Further, weaning weight could be considered as desirable trait for earliest indirect selection for wool yield in view of its high genetic correlation with wool traits.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.9 no.5
• /
• pp.986-994
• /
• 2005

This Paper proposes the likelihood-based nonlinear dimension reduction method of the speech feature parameters in order to construct the voice activity detecter adaptable in noisy environment. The proposed method uses the nonlinear values of the Gaussian probability density function with the new parameters for the speec/nonspeech class. We adapted Likelihood Ratio Test to find speech part and compared its performance with that of Linear Discriminant Analysis technique. In experiments we found that the proposed method has the similar results to that of Gaussian Mixture Models.

• Communications for Statistical Applications and Methods
• /
• v.15 no.5
• /
• pp.727-735
• /
• 2008

Consider a clinical trial in which the main end-point is survival. Suppose after the start of the study an intermediate event occurs which may be influenced by a covariate(or treatment). In many clinical studies the occurrence of an intermediate event may change the survival distribution. This investigation develops two-stage model which, in the first stage, models the effect of covariate on the intermediate event and models the relationship between survival time and covariate as well as the intermediate event. In this paper, the two-stage model is presented in order to model intermediate event and a test based on this model is also provided. A numerical simulations are carried out to evaluate its overall significance level.

• Journal of Korean Society for Quality Management
• /
• v.25 no.3
• /
• pp.96-107
• /
• 1997

The recent stream to reliability prediction is that it is totally inclusive in depth to consider even the operating and environmental condition at the level of finished goods as well as component itselves. In this study, firstly we present the reliability prediction methods by entire failure rate model which failure rate at the system level is added to the failure rate model at the component level. Secondly we build up the improved bases of reliability demonstration through a, pp.ication of Kaplan-Meier, Cumulative hazard, Johnson's methods as non-parametric and Maximum Likelihood Estimator under exponential & Weibull distribution as parametric. And also present the methods of curve fitting to piecewise failure rate under Weibull distribution, PRST (Probability Ratio Sequential Test), curve fitting to S-shaped reliability growth curve, computer programs of each methods. Lastly we show the practical for determination of optimal burn-in time as a method of reliability enhancement, and also verify the practical usefulness of the above study through the a, pp.ication of failure and test data during 1 year.

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.6
• /
• pp.1661-1671
• /
• 2016

Risk analysis is a systematic study of uncertainties and risks we encounter in business, engineering, public policy, and many other areas. Value at Risk (VaR) is one of the most widely used risk measurements in risk management. In this paper, the Korean Composite Stock Price Index data has been utilized to model the VaR employing the classical ARMA (1,1)-GARCH (1,1) models with normal, t, generalized hyperbolic, and generalized pareto distributed errors. The aim of this paper is to compare the performance of each model in estimating the VaR. The performance of models were compared in terms of the number of VaR violations and Kupiec exceedance test. The GARCH-GPD likelihood ratio unconditional test statistic has been found to have the smallest value among the models.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.59 no.9
• /
• pp.1673-1679
• /
• 2010

A photovoltaic (PV) generator is significantly regarded as one important alternative of renewable energy systems recently. Fault detection and diagnosis of engineering dynamic systems is a fundamental issue to timely prevent unexpected damages in industry fields. This paper presents an intelligent monitoring approach and fault detection technique for PV generator systems by means of artificial neural network and statistical signal detection theory. We devise a multi-Fourier neural network model for representing dynamics of PV systems and apply a general likelihood ratio test (GLRT) approach for investigating our decision making algorithm in fault detection and diagnosis. We make use of a test-bed of ubiquitous sensor network (USN) based PV monitoring systems for testing our proposed fault detection methodology. Lastly, a real-time experiment is accomplished for demonstrating its reliability and practicability.