• The Korean Journal of Applied Statistics
• /
• v.27 no.6
• /
• pp.909-922
• /
• 2014

This paper reviews Bayesian inference with inequality constraints. It focuses on ⅰ) comparison of models with various inequality/equality constraints on parameters, ⅱ) multiple tests on equalities of parameters when parameters are under inequality constraints, ⅲ) multiple test on equalities of score parameters in models for contingency tables with ordinal categorical variables.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1995.06a
• /
• pp.83-87
• /
• 1995

Keyword Spotting 이란 음성인식의 한 분야로서 입력된 음성에서 미리 정해진 특정단어 또는 복수 개의 단어들 중 어느 것이 포함되어 있는지의 여부를 찾아내고 이 단어를 식별해 내는 작업을 의미한다. 음소모델을 이용하여 Keyword Spotting 시스템을 구성할 경우 새로운 keyword의 추가 또는 변경이 필요할 때 단순히 그 발음사전에 따라 음소모델들을 연결시킴으로써 keyword 모델을 구성할 수 있으므로 단어모델에 의한 방법에 비해 장점이 있다. 본 논문에서는 triphone을 기본단위로 하는 HMM 에 의해 keyword 모델을 구성하고, non-keyword 모델 및 silence 모델을 함께 사용하는 keyword spotting 시스템을 구성하였다. 이러한 시스템에서 non-keyword 모델은 keyword와 keyword가 아닌 음성을 구분 지어주는 역할을 하므로 인식성능의 향상을 위해서는 적절한 non-keyword 모델의 선택이 필요하다. 본 논문에서는 10개의 state를 갖는 단일모델, 조음방법에 의해 음소들을 clustering 한 모델, 그리고 통계적 방법에 의해 음소들을 clustering 한 모델들을 각각 non-keyword 모델로 사용하여 그 성능을 비교하였다. 6개의 keyword를 대상으로 한 화자독립 keyword spotting 실험결과, 통계적 방법에 의해 음소들을 6 또는 7개의 그룹으로 clustering 한 방법이 가장 우수한 인식성능을 나타냈다.

• Journal of the Korean Data and Information Science Society
• /
• v.10 no.1
• /
• pp.119-133
• /
• 1999

A Markov Chain Monte Carlo method is developed to compute the software reliability model. We consider computation problem for determining of posterior distibution in Bayseian inference. Metropolis algorithms along with Gibbs sampling are proposed to preform the Bayesian inference of the Mixed model with record value statistics. For model determiniation, we explored the prequential conditional predictive ordinate criterion that selects the best model with the largest posterior likelihood among models using all possible subsets of the component intensity functions. To relax the monotonic intensity function assumptions. A numerical example with simulated data set is given.

• Speech Sciences
• /
• v.11 no.4
• /
• pp.75-88
• /
• 2004

In this paper, we propose a new approach to sequential linear regression adaptation of continuous density hidden Markov models (CDHMMs) based on transformation space model (TSM). The proposed TSM which characterizes the a priori knowledge of the training speakers associated with maximum likelihood linear regression (MLLR) matrix parameters is effectively described in terms of the latent variable models. The TSM provides various sources of information such as the correlation information, the prior distribution, and the prior knowledge of the regression parameters that are very useful for rapid adaptation. The quasi-Bayes (QB) estimation algorithm is formulated to incrementally update the hyperparameters of the TSM and regression matrices simultaneously. Experimental results showed that the proposed TSM approach is better than that of the conventional quasi-Bayes linear regression (QBLR) algorithm for a small amount of adaptation data.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.3
• /
• pp.852-861
• /
• 2000

Software reliability growth models are used in testing stages of software development to model the error content and time intervals betwewn software failures. In this paper, using priors for the number of fault with the negative binomial distribution nd the error rate with gamma distribution, Bayesian inference and model selection method for Jelinski-Moranda and Goel-Okumoto and Schick-Wolverton models in software reliability. For model selection, we explored the sum of the relative error, Braun statistic and median variation. In Bayesian computation process, we could avoid the multiple integration by the use of Gibbs sampling, which is a kind of Markov Chain Monte Carolo method to compute the posterior distribution. Using simulated data, Bayesian inference and model selection is studied.

• Communications for Statistical Applications and Methods
• /
• v.22 no.4
• /
• pp.349-359
• /
• 2015

We study a semiparametric Bayesian approach to small area estimation under a nested error linear regression model with area level covariate subject to measurement error. Consideration is given to radial basis functions for the regression spline and knots on a grid of equally spaced sample quantiles of covariate with measurement errors in the nested error linear regression model setup. We conduct a hierarchical Bayesian structural measurement error model for small areas and prove the propriety of the joint posterior based on a given hierarchical Bayesian framework since some priors are defined non-informative improper priors that uses Markov Chain Monte Carlo methods to fit it. Our methodology is illustrated using numerical examples to compare possible models based on model adequacy criteria; in addition, analysis is conducted based on real data.

• The Bulletin of The Korean Astronomical Society
• /
• v.44 no.1
• /
• pp.41.1-41.1
• /
• 2019

We consider dynamical dark energy models based on a minimally coupled scalar field with three different potentials: the inverse power-law, SUGRA and double exponential potentials. For each model, we derived perturbation initial conditions in the early epoch and performed the Markov Chain Monte Carlo (MCMC) analysis to explore the parameter space that is favored by the current cosmological observations like Planck CMB anisotropy, type Ia supernovae, and baryon acoustic oscillation data. The analysis has been done by using the modified CAMB/COSMOMC code in which the dynamical evolution of the scalar field perturbations are fully considered. The MCMC constraints on the cosmological as well as potential parameters are derived. In the talk we will present a progress report.

• Communications for Statistical Applications and Methods
• /
• v.28 no.2
• /
• pp.99-118
• /
• 2021

In this paper, we introduce an extended form of the inverse power Lomax model via Marshall-Olkin approach. We call it the Marshall-Olkin inverse power Lomax (MOIPL) distribution. The four- parameter MOIPL distribution is very flexible which contains some former and new models. Vital properties of the MOIPL distribution are affirmed. Maximum likelihood estimators and approximate confidence intervals are considered under Type I censored samples. Maximum likelihood estimates are evaluated according to simulation study. Bayesian estimators as well as Bayesian credible intervals under symmetric loss function are obtained via Markov chain Monte Carlo (MCMC) approach. Finally, the flexibility of the new model is analyzed by means of two real data sets. It is found that the MOIPL model provides closer fits than some other models based on the selected criteria.

• Nuclear Engineering and Technology
• /
• v.53 no.2
• /
• pp.357-367
• /
• 2021

In general, common cause failures (CCFs) have been modeled with the assumption that components within the same group are symmetric. This assumption reduces the number of parameters required for the CCF probability estimation and allows us to use a parametric model, such as the alpha factor model. Although there are various asymmetric conditions in nuclear power plants (NPPs) to be addressed, the traditional CCF models are limited to symmetric conditions. Therefore, this paper proposes the copulabased CCF model to deal with asymmetric as well as symmetric CCFs. Once a joint distribution between the components is constructed using copulas, the proposed model is able to provide the probability of common cause basic events (CCBEs) by formulating a system of equations without symmetry assumptions. In addition, Bayesian inferences for the parameters of the marginal and copula distributions are introduced and Markov Chain Monte Carlo (MCMC) algorithms are employed to sample from the posterior distribution. Three example cases using simulated data, including asymmetry conditions in total failure probabilities and/or dependencies, are illustrated. Consequently, the copula-based CCF model provides appropriate estimates of CCFs for asymmetric conditions. This paper also discusses the limitations and notes on the proposed method.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.12
• /
• pp.2647-2654
• /
• 2011

In this paper, the algorithm that recognizes the gesture by configuring the feature information obtained through edge orientation histogram and principal component analysis as low dimensional gesture symbol was described. Since the proposed method doesn't require a lot of computations compared to the existing geometric feature based method or appearance based methods and it can maintain high recognition rate by using the minimum information, it is very well suited for real-time system establishment. In addition, to reduce incorrect recognition or recognition errors that occur during gesture recognition, the model feature values projected in the gesture space is configured as a particular status symbol through clustering algorithm to be used as input symbol of hidden Markov models. By doing so, any input gesture will be recognized as the corresponding gesture model with highest probability.