• Proceedings of the Korean Information Science Society Conference
• /
• 2005.07b
• /
• pp.628-630
• /
• 2005

베이지안 네트워크는 확률이론에 기초해 불확실성이 존재하는 실세계 문제를 해결하는데 많은 기여를 하고 있다. 최근 네트워크 구조를 데이터로부터 자동으로 학습하는 많은 연구가 이루어져 보다 손쉽게 많은 사람들이 사용할 수 있게 되었다. 하지만 한번 학습하여 고정된 네트워크의 구조는 새롭게 수집되는 데이터의 특성을 잘 반영하지 못하는 문제를 지니고 있다. 환경의 변화에 맞게 지속적으로 네트워크 구조를 갱신하기 위한 연구가 진행되고 있으며 본 연구에서는 Lam이 제안한 MDL기반 평가함수를 이용한 진화적 갱신 방법을 제안하여 갱신 성능을 향상시키고자 한다. 벤치마크 네트워크인 ASIA에 대한 실험 결과 제안한 방법이 기존의 지역적 탐색 방법에 비해 향상된 성능을 제공함을 확인하였다.

• Journal of Navigation and Port Research
• /
• v.33 no.1
• /
• pp.9-19
• /
• 2009

This paper describes empirical approach methodology for the quantitative risk assessment of maritime transportation accident (MTA) of a merchant ship. The principal aim of this project is to estimate the risk of MTA that could degrade the ship safety by analyzing the underlying factors contributing to MTA based on the IMO's Formal Safety Assessment techniques and, by assessing the probabilistic risk level of MTA based on the quantitative risk assessment methodology. The probabilistic risk level of MTA to Risk Index (RI) composed with Probability Index (PI) and Severity Index (SI) can be estimated from proposed Maritime Transportation Accident Model (MTAM) based on Bayesian Network with Bayesian theorem Then the applicability of the proposed MTAM can be evaluated using the scenario group with 355 core damaged accident history. As evaluation results, the correction rate of estimated PI, $r_{Acc}$ is shown as 82.8%, the over ranged rate of PI variable sensitivity with $S_p{\gg}1.0$ and $S_p{\ll}1.0$ is shown within 10%, the averaged error of estimated SI, $\bar{d_{SI}}$ is shown as 0.0195 and, the correction rate of estimated RI, $r_{Acc}$(%), is shown as 91.8%. These results clearly shown that the proposed accident model and methodology can be use in the practical maritime transportation field.

• Journal of the Korean Geotechnical Society
• /
• v.26 no.7
• /
• pp.161-170
• /
• 2010

For the development of load and resistance factor design, reliability analysis is required to calibrate resistance factors in the framework of reliability theory. The distribution of measured-to-predicted pile resistance ratio was obrained based on only the results of load tests conducted to failure for the assessment of uncertainty regarding pile resistance and used in the conventional reliability analysis. In other words, successful pile load test (piles resisted twice their design loads without failure) results were discarded, and therefore, were not reflected in the reliability analysis. In this paper, a new systematic method based on Bayesian theory is used to update reliability indices of driven steel pipe piles by adding more proof pile load test results, even not conducted to failure, to the prior distribution of pile resistance ratio. Fifty seven static pile load tests performed to failure in Korea were compiled for the construction of prior distribution of pile resistance ratio. The empirical method proposed by Meyerhof is used to calculate the predicted pile resistance. Reliability analyses were performed using the updated distribution of pile resistance ratio. The challenge of this study is that the distribution updates of pile resistance ratio are possible using the load test results even not conducted to failure, and that Bayesian updates are most effective when limited data are available for reliability analysis.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.25 no.4
• /
• pp.355-362
• /
• 2012

Remaining useful life(RUL) prediction of a system is important in the prognostics field since it is directly linked with safety and maintenance scheduling. In the physics-based prognostics, accurately estimated model parameters can predict the remaining useful life exactly. It, however, is not a simple task to estimate the model parameters because most real system have multivariate model parameters, also they are correlated each other. This paper presents representative methods to estimate model parameters in the physics-based prognostics and discusses the difference between three methods; the particle filter method(PF), the overall Bayesian method(OBM), and the sequential Bayesian method(SBM). The three methods are based on the same theoretical background, the Bayesian estimation technique, but the methods are distinguished from each other in the sampling methods or uncertainty analysis process. Therefore, a simple physical model as an easy task and the Paris model for crack growth problem are used to discuss the difference between the three methods, and the performance of each method evaluated by using established prognostics metrics is compared.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.406-406
• /
• 2021

극한강우사상의 분석은 다양한 극치 분포로 구성된 극치이론을 통해 가능하다. 일반적으로 단일 지점의 극한사상의 분석을 위한 지점빈도해석 (Point Frequency Analysis, PFA)이 다양한 재현기간에 해당하는 강우량을 추정하는데 널리 사용되어왔다. 하지만 수문기후학적 극치기록은 시간적 그리고 공간적으로 제한적이다. 따라서 모의 불확실성을 줄이고 신뢰성 높은 결과를 도출하기 위해 서로 유사한 분포를 가질 수 있는 인근 지점의 활용하는 지역빈도해석 (Regional Frequency Analysis, RFA) 방법이 개발되어 적용되고 있다. 본 연구에서는 부산, 울산, 경남지역의 기상청 종관기상관측시스템(Automated Synoptic Observing System, ASOS) 울산, 부산, 통영, 진주, 거창, 합천, 밀양, 산청, 거제, 남해지점 일강수량을 자료를 기반으로 Metropolis-Hasting 알고리즘을 사용하여 일반극치분포(Generalized Extreme Value, GEV)의 매개변수를 추정하고 PFA 및 RFA의 불확실성을 평가하고자 한다. 이러한 연구는 공간적 구성 요소(예, 지리적 좌표, 고도)를 고려하지 못하며 추가변수 (예, 공변량)를 분석에 결합할 수 없는 등의 RFA의 한계를 극복하고, 명시적으로 불확실성을 추정하여 결과의 신뢰성을 확보 할 수 있는 계층적 베이지안 모델의 개발에 도움이 되리라 기대된다.

• The Korean Journal of Applied Statistics
• /
• v.24 no.5
• /
• pp.833-845
• /
• 2011

This paper conducts a statistical analysis of extreme values for both daily log-returns and daily negative log-returns, which are computed using a collection of KOSPI data from January 3, 1998 to August 31, 2011. The Poisson-GPD model is used as a statistical analysis model for extreme values and the maximum likelihood method is applied for the estimation of parameters and extreme quantiles. To the Poisson-GPD model is also added the Bayesian method that assumes the usual noninformative prior distribution for the parameters, where the Markov chain Monte Carlo method is applied for the estimation of parameters and extreme quantiles. According to this analysis, both the maximum likelihood method and the Bayesian method form the same conclusion that the distribution of the log-returns has a shorter right tail than the normal distribution, but that the distribution of the negative log-returns has a heavier right tail than the normal distribution. An advantage of using the Bayesian method in extreme value analysis is that there is nothing to worry about the classical asymptotic properties of the maximum likelihood estimators even when the regularity conditions are not satisfied, and that in prediction it is effective to reflect the uncertainties from both the parameters and a future observation.

• The KIPS Transactions:PartB
• /
• v.11B no.4
• /
• pp.485-490
• /
• 2004

In data representation, the clustering performs a grouping process which combines given data into some similar clusters. The various similarity measures have been used in many researches. But, the validity of clustering results is subjective and ambiguous, because of difficulty and shortage about objective criterion of clustering. The fuzzy clustering provides a good method for subjective clustering problems. It performs clustering through the similarity matrix which has fuzzy membership value for assigning each object. In this paper, for objective fuzzy clustering, the clustering algorithm which joins principal components analysis as a dimension reduction model with bayesian learning as a statistical learning theory. For performance evaluation of proposed algorithm, Iris and Glass identification data from UCI Machine Learning repository are used. The experimental results shows a happy outcome of proposed model.

• Journal of KIISE:Computer Systems and Theory
• /
• v.33 no.12
• /
• pp.885-893
• /
• 2006

The Bayesian network(BN) has emerged in recent years as a powerful technique for handling uncertainty iii complex domains. Parameter learning of BN to find the most proper network from given data set has been investigated to decrease the time and effort for designing BN. Off-line learning needs much time and effort to gather the enough data and since there are uncertainties in real world, it is hard to get the complete data. In this paper, we propose an online learning method of Bayesian network parameters from incomplete data. It provides higher flexibility through learning from incomplete data and higher adaptability on environments through online learning. The results of comparison with Voting EM algorithm proposed by Cohen at el. confirm that the proposed method has the same performance in complete data set and higher performance in incomplete data set, comparing with Voting EM algorithm.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.38B no.7
• /
• pp.543-552
• /
• 2013

For the safety messages in IEEE 802.11p/WAVE vehicles network environment, strict periodic beacon broadcasting requires status advertisement to assist the driver for safety. In crowded networks where beacon message are broadcasted at a high number of frequencies by many vehicles, which used for beacon sending, will be congested by the wireless medium due to the contention-window based IEEE 802.11p MAC. To resolve the congestion, we consider a MAC scheme based on slotted p-persistent CSMA as a simple non-cooperative Bayesian game which involves payoffs reflecting the attempt probability. Then, we derive Bayesian Nash Equilibrium (BNE) in a closed form. Using the BNE, we propose new congestion control algorithm to improve the performance of the beacon rate under saturation condition in IEEE 802.11p/WAVE vehicular networks. This algorithm explicitly computes packet delivery probability as a function of contention window (CW) size and number of vehicles. The proposed algorithm is validated against numerical simulation results to demonstrate its stability.

• Economic Analysis
• /
• v.26 no.1
• /
• pp.84-132
• /
• 2020

A central bank infers market expectations of future yields based on yield curves. The central bank needs to precisely understand the changes in market expectations of future yields in order to have a more effective monetary policy. This need explains why a range of models have attempted to produce yield curves and market expectations that are as accurate as possible. Alongside the development of bond markets, the interconnectedness between them and macroeconomic factors has deepened, and this has rendered understanding of what macroeconomic variables affect yield curves even more important. However, the existence of various theories about determinants of yields inevitably means that previous studies have applied different macroeconomics variables when estimating yield curves. This indicates model uncertainties and naturally poses a question: Which model better estimates yield curves? Put differently, which variables should be applied to better estimate yield curves? This study employs the Dynamic Nelson-Siegel Model and takes the Bayesian approach to variable selection in order to ensure precision in estimating yield curves and market expectations of future yields. Bayesian variable selection may be an effective estimation method because it is expected to alleviate problems arising from a priori selection of the key variables comprising a model, and because it is a comprehensive approach that efficiently reflects model uncertainties in estimations. A comparison of Bayesian variable selection with the models of previous studies finds that the question of which macroeconomic variables are applied to a model has considerable impact on market expectations of future yields. This shows that model uncertainties exert great influence on the resultant estimates, and that it is reasonable to reflect model uncertainties in the estimation. Those implications are underscored by the superior forecasting performance of Bayesian variable selection models over those models used in previous studies. Therefore, the use of a Bayesian variable selection model is advisable in estimating yield curves and market expectations of yield curves with greater exactitude in consideration of the impact of model uncertainties on the estimation.