• Journal of the Korean Statistical Society
• /
• 제29권2호
• /
• pp.155-168
• /
• 2000

Regarding to multiple comparison problem (MCP) of k normal population variances, we suggest a Bayesian method for calculating posterior probabilities for various hypotheses of equality among population variances. This leads to a simple method for obtaining pairwise comparisons of variances in a statistical experiment with a partition on the parameter space induced by equality and inequality relationships among the variances. The method is derived from the fact that certain features of the hierarchical nonparametric family of Dirichlet process priors, in general, make it amenable to solving the MCP and estimating the posterior probabilities by means of posterior simulation, the Gibbs sampling. Two examples are illustrated for the method. For these examples, the method is straightforward for specifying distributionally and to implement computationally, with output readily adapted for required comparison.

• Journal of the Korean Data and Information Science Society
• /
• 제16권4호
• /
• pp.1129-1140
• /
• 2005

A nonparametric Bayesian method for calculating posterior probabilities of the multiple comparison problem on the parameters of several Geometric populations is presented. Bayesian multiple comparisons under two different prior/ likelihood combinations was studied by Gopalan and Berry(1998) using Dirichlet process priors. In this paper, we followed the same approach to calculate posterior probabilities for various hypotheses in a statistical experiment with a partition on the parameter space induced by equality and inequality relationships on the parameters of several geometric populations. This also leads to a simple method for obtaining pairwise comparisons of probability of successes. Gibbs sampling technique was used to evaluate the posterior probabilities of all possible hypotheses that are analytically intractable. A numerical example is given to illustrate the procedure.

• Communications for Statistical Applications and Methods
• /
• 제8권2호
• /
• pp.427-434
• /
• 2001

In this paper, we consider the nonparametric Bayesian approach to the multiple comparisons problem for I Poisson populations using Dirichlet process priors. We describe Gibbs sampling algorithm for calculating posterior probabilities for the hypotheses and calculate posterior probabilities for the hypotheses using Markov chain Monte Carlo. Also we provide a numerical example to illustrate the developed numerical technique.

• Spatial Information Research
• /
• 제12권1호
• /
• pp.111-125
• /
• 2004

위성영상 분류에 관한 주요 주제 중 하나는 분류 정확도 향상에 있다. 동일지역에 대한 동일시기의 위성영상을 취득할 수 있는 기회가 많아지는 현실을 감안할 때, 복수의 위성영상 데이터를 이용하여 분류정확도가 향상된 분류결과를 도출하는 것은 의미 있는 일일 것이다. 본 연구 주제는 최대우도법을 사용하여 계산된 데이터의 사후확률 및 분류 불확실도를 Dempster-Shafer의 증거이론에 적용하여 분류정확도를 향상시키고자 하는 것이다. 분석결과 개별적인 데이터 분류나 데이터간 융합에 의한 분류보다 본 연구에서 제안한 방법이 전체정확도와 Kappa 지수 모두 높은 정확도를 나타냈으며, 정확도 차에 대한 검정을 실시하여 본 연구에서 제안한 방법이 다른 방법에 비해 우수함을 통계적으로 증명하였다.

• 응용통계연구
• /
• 제15권1호
• /
• pp.73-84
• /
• 2002

계수에 대한 부등 제한조건이 있는 선형 회귀모형은 경제모형에서 가장 흔하게 다루어지는 것 중의 하나이다. 이는 특정 설명변수에 대한 계수의 부호를 음양 중 하나로 제한하거나 계수들에 대하여 순서적 관계를 주기 때문이다. 본 논문에서는 이러한 부등 제한이 있는 선형회귀 모형에서 유의한 설명변수의 선택을 해결하는 베이지안 기법을 고려한다. 베이지안 변수선택은 가능한 모든 모형의 사후확률 계산이 요구되는데 본 논문에서는 이러한 사후확률들을 동시에 계산하는 방법을 제시한다. 구체적으로 가장 일반적인 모형의 모수에 대한 사후표본을 깁스 표본기법을 적용시켜 얻은 후 이를 이용하여 모든 가능한 모형의 사후확률을 계산하고 실제적인 자료에 본 논문에서 제안된 방법을 적용시켜 본다.

• 한국산업정보학회:학술대회논문집
• /
• 한국산업정보학회 1999년도 춘계학술대회 발표논문집
• /
• pp.191-197
• /
• 1999

In this paper, we develop noninformative priors that are used for estimating the reliability of stress-strength system under the Burr-type X distribution. A class of priors is found by matching the coverage probabilities of one-sided Bayesian credible interval with the corresponding frequentist coverage probabilities. It turns out that the reference prior is a first order matching prior. The propriety of posterior under matching prior is provided. The frequentist coverage probabilities are given for small samples.

• 한국멀티미디어학회논문지
• /
• 제23권7호
• /
• pp.883-890
• /
• 2020

In this paper, we propose a variational expectation-maximization algorithm that computes posterior probabilities from Latent Dirichlet Allocation (LDA) model. The algorithm approximates the intractable posterior distribution of a document term matrix generated from a corpus made up by 50 papers. It approximates the posterior by searching the local optima using lower bound of the true posterior distribution. Moreover, it maximizes the lower bound of the log-likelihood of the true posterior by minimizing the relative entropy of the prior and the posterior distribution known as KL-Divergence. The experimental results indicate that documents clustered to image classification and segmentation are correlated at 0.79 while those clustered to object detection and image segmentation are highly correlated at 0.96. The proposed variational inference algorithm performs efficiently and faster than Gibbs sampling at a computational time of 0.029s.

• 한국멀티미디어학회논문지
• /
• 제23권4호
• /
• pp.595-602
• /
• 2020

We propose a generative probabilistic model with Dirichlet prior distribution for topic modeling and text similarity analysis. It assigns a topic and calculates text correlation between documents within a corpus. It also provides posterior probabilities that are assigned to each topic of a document based on the prior distribution in the corpus. We then present a Gibbs sampling algorithm for inference about the posterior distribution and compute text correlation among 50 abstracts from the papers published by IEEE. We also conduct a supervised learning to set a benchmark that justifies the performance of the LDA (Latent Dirichlet Allocation). The experiments show that the accuracy for topic assignment to a certain document is 76% for LDA. The results for supervised learning show the accuracy of 61%, the precision of 93% and the f1-score of 96%. A discussion for experimental results indicates a thorough justification based on probabilities, distributions, evaluation metrics and correlation coefficients with respect to topic assignment.

• 한국컴퓨터정보학회논문지
• /
• 제23권6호
• /
• pp.1-7
• /
• 2018

A multi-label classification is to find multiple labels associated with the input pattern. Multi-label classification can be achieved by extending conventional single-label classification. Common extension techniques are known as Binary relevance, Label powerset, and Classifier chains. However, most of the extended multi-label naive bayes classifier has not been able to accurately estimate posterior probabilities because it does not reflect the label dependency. And the remaining extended multi-label naive bayes classifier has a problem that it is unstable to estimate posterior probability according to the label selection order. To estimate posterior probability well, we propose a new posterior probability estimation method that reflects the probability between all labels and labels efficiently. The proposed method reflects the correlation between labels. And we have confirmed through experiments that the extended multi-label naive bayes classifier using the proposed method has higher accuracy then the existing multi-label naive bayes classifiers.

• Journal of the Korean Statistical Society
• /
• 제29권1호
• /
• pp.95-108
• /
• 2000

Consider the oblique factor model X=Af＋$\varepsilon$, with defining relation $\Sigma$=Λ$\Phi$Λ'＋Ψ. This paper is concerned with suggesting an optimal Bayes criterion for determining the number of factors in the model, i.e. dimension of the vector f. The use of marginal likelihood as a method for calculating posterior probability of each model with given dimension is developed under a generalized conjugate prior. Then based on an appropriate loss function, a Bayes rule is developed by use of the posterior probabilities. It is shown that the approach is straightforward to specify distributionally and to imploement computationally, with output readily adopted for constructing required cirterion.