• Proceedings of the Korea Water Resources Association Conference
• /
• 2019.05a
• /
• pp.338-338
• /
• 2019

설계강우량 산정시, 일반적으로 극치자료를 활용하여 정상성 가정하에 빈도해석을 수행하고 있다. 그러나 종종 정상성으로 가정했던 기존 극치강우자료가 정상성 빈도해석 모형에서 효과적으로 모델링되지 않는 비정상성 특성을 나타내고 있다. 또한, 대부분의 극치강우 분포는 해마다 다른 규모로 발생하는 홍수와 태풍 등의 강우요인으로 인해 두 개의 첨두를 갖는 혼합분포 형태를 보인다. 이에 본 연구에서는 혼합분포 기반 비정상성 빈도모델(mixture distribution based nonstationary frequency model, MDNF)을 제시하였다. 제안된 모형의 입력자료로 기후변수(e.g. SSTs and SLPs)를 사용하여 두 개의 분포형으로 구성되는 극치강우의 혼합비(mixing ratio)에 대한 영향을 분석하였으며, 극치강우 패턴이 특정 기후변수의 영향을 받는 것을 확인하였다. 최종적으로 Bayesian 기법을 MDNF 모형에 연계하여 각 첨두에 해당하는 분포형의 매개변수들에 대한 불확실성 구간을 정량적으로 제시하였다. 본 연구를 통해 강우 패턴의 변동은 설계 강우량 추정에 영향을 미치며, 특정 기후변수와 강우 패턴이 상관성을 가지는 것을 확인함으로써 합리적인 설계 강우량 산정을 위한 중요한 근거를 제공할 것으로 사료된다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.6
• /
• pp.796-801
• /
• 2009

In medical practice, the diagnosis or prediction models requiring complicated computations are not widely recognized due to difficulty in interpreting the course of reasoning and the complexity of computations. Medical personnel have used the nomograms which are a graphical representation for numerical relationships that enables to easily compute a complicated function without help of computation machines. It has been widely paid attention in diagnosing diseases or predicting the progress of diseases. A nomogram is constructed from a set of clinical data which contain various attributes such as symptoms, lab experiment results, therapy history, progress of diseases or identification of diseases. It is of importance to select effective ones from available attributes, sometimes along with parameters accompanying the attributes. This paper introduces a nomogram construction method that uses a naive Bayesian technique to construct a nomogram as well as a genetic algorithm to select effective attributes and parameters. The proposed method has been applied to the construction of a nomogram for a real clinical data set.

• Journal of Coastal Disaster Prevention
• /
• v.4 no.3
• /
• pp.119-131
• /
• 2017

Global warming under the influence of climate change and its direct impact on glacial and sea level are known issue. However, there is a lack of research on an indirect impact of climate change such as coastal structure design which is mainly based on a frequency analysis of water level under the stationary assumption, meaning that maximum sea level will not vary significantly over time. In general, stationary assumption does not hold and may not be valid under a changing climate. Therefore, this study aims to develop a novel approach to explore possible distributional changes in annual maximum sea levels (AMSLs) and provide the estimate of design water level for coastal structures using a multiple non-crossing quantile regression based nonstationary frequency analysis within a Bayesian framework. In this study, 20 tide gauge stations, where more than 30 years of hourly records are available, are considered. First, the possible distributional changes in the AMSLs are explored, focusing on the change in the scale and location parameter of the probability distributions. The most of the AMSLs are found to be upward-convergent/divergent pattern in the distribution, and the significance test on distributional changes is then performed. In this study, we confirm that a stationary assumption under the current climate characteristic may lead to underestimation of the design sea level, which results in increase in the failure risk in coastal structures. A detailed discussion on the role of the distribution changes for design water level is provided.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.18 no.6A
• /
• pp.129-137
• /
• 2008

As the use of the internet increases, the distribution of spam mail has also vastly increased. The email's main use was for the exchange of information, however, currently it is being more frequently used for advertisement and malware distribution. This is a serious problem because it consumes a large amount of the limited internet resources. Furthermore, an extensive amount of computer, network and human resources are consumed to prevent it. As a result much research is being done to prevent and filter spam. Currently, research is being done on readable sentences which do not use proper grammar. This type of spam can not be classified by previous vocabulary analysis or document classification methods. This paper proposes a method to filter spam by using the subject of the mail and N-GRAM for indexing and Bayesian, SVM algorithms for classification.

• Journal of the Semiconductor & Display Technology
• /
• v.22 no.1
• /
• pp.134-138
• /
• 2023

This study focuses on the analysis of the results of computational fluid dynamics simulations of mist-chemical vapor deposition for the growth of an epitaxial wafer in power semiconductor technology using artificial intelligence techniques. The conventional approach of predicting the uniformity of the deposited layer using computational fluid dynamics and design of experimental takes considerable time. To overcome this, artificial intelligence method, which is widely used for optimization, automation, and prediction in various fields, was utilized to analyze the computational fluid dynamics simulation results. The computational fluid dynamics simulation results were analyzed using a supervised deep neural network model for regression analysis. The predicted results were evaluated quantitatively using Euclidean distance calculations. And the Bayesian optimization was used to derive the optimal condition, which results obtained through deep neural network training showed a discrepancy of approximately 4% when compared to the results obtained through computational fluid dynamics analysis. resulted in an increase of 146.2% compared to the previous computational fluid dynamics simulation results. These results are expected to have practical applications in various fields.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.37 no.9
• /
• pp.1133-1140
• /
• 2013

A computer model is a useful tool that provides solution via physical modeling instead of expensive testing. In reality, however, it often does not agree with the experimental data owing to simplifying assumption and unknown or uncertain input parameters. In this study, a Bayesian approach is proposed to calibrate the computer model in a probabilistic manner using the measured data. The elasto-plastic analysis of a pyrotechnically actuated device (PAD) is employed to demonstrate this approach, which is a component that delivers high power in remote environments by the combustion of a self-contained energy source. A simple mathematical model that quickly evaluates the performance is developed. Unknown input parameters are calibrated conditional on the experimental data using the Markov Chain Monte Carlo algorithm, which is a modern computational statistics method. Finally, the results are applied to determine the reliability of the PAD.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.04c
• /
• pp.389-391
• /
• 2003

컴퓨터 학습의 군집화는 주어진 데이터를 서로 유사한 몇 개의 집단으로 묶는 작업을 수행한다. 군집화를 위한 유사도 결정을 위한 측도는 많은 기법들에서 매우 다양한 측도들이 사용되고 또한 연구되어 왔다. 하지만 군집화의 결과에 대한 성능측정에 대한 객관적인 기준 설정이 어렵기 때문에 군집화 결과에 대한 해석은 매우 주관적이고 애매한 경우가 많다. 퍼지 군집화는 이러한 애매한 군집화 문제에 있어서 융통성 있는 군집 결정 방안을 제시해 준다. 각 개체들이 특정 군집에 속하게 될 퍼지 멤버 함수값을 원소로 하는 유사도 행렬을 통하여 군집화를 수행한다. 본 논문에서는 베이지안 학습을 통하여 군집화를 위한 퍼지 멤버 함수값을 구하였다. 본 연구에서는 최적의 퍼지 군집화 수행을 위하여 베이지안 학습 기반의 퍼지 규칙을 추출하였다. 인공적으로 만든 데이터와 기존의 기계 학습 데이터를 이용한 실험을 통하여 제안 방법의 성능을 확인하였다.

• Proceedings of the Korean Statistical Society Conference
• /
• 2002.11a
• /
• pp.61-68
• /
• 2002

통계학은 불확실성(uncertainty)에 대한 연구이다. 베이지안 통계 방법은 불확실성 아래서 통계 추론과 의사 결정 모두를 위한 완전한(complete) 패러다임을 제공한다. 베이지안 방법론은 합리적인 초기 정보와 결합하는 것을 가능하게 만들고, 전통적인 통계적 방법론에 의하여 직면하는 많은 어려움들을 풀 수 있는 coherent 방법론을 제공하면서 엄격한 수학적 기본에 근거하고 있다. 베이지안 패러다임은 일반적인 용어로써 확률이란 단어의 사용을 가장 잘 어울리게 하는 불확실성의 조건부 측도(conditional measure of uncertainty)로써 확률의 해석에 근거한다. 관심있는 것에 대한 통계적 추론은 증거의 관점에서 그 값에 대한 불확실성의 변형으로써 묘사되며, 베이즈 정리(Bayes' theorem)는 이러한 변형이 어떻게 만들어지는 가를 자세히 설명할 수 있다. 베이지안 방법들은 전통적인 통계적 방법론에 접근할 없는 복잡하고, 다양한 구조적 문제들에 응용할 수 있다.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.43 no.2
• /
• pp.69-77
• /
• 2001

An artificial neural network model was developed to analyze and forecast daily steamflow flow a small watershed. Error Back propagation neural networks (EBPN) of daily rainfall and runoff data were found to have a high performance in simulating stremflow. The model adopts a gradient descent method where the momentum and adaptive learning rate concepts were employed to minimize local minima value problems and speed up the convergence of EBP method. The number of hidden nodes was optimized using Bayesian information criterion. The resulting optimal EBPN model for forecasting daily streamflow consists of three rainfall and four runoff data (Model34), and the best number of the hidden nodes were found to be 13. The proposed model simulates the daily streamflow satisfactorily by comparison compared to the observed data at the HS#3 watershed of the Baran watershed project, which is 391.8 ha and has relatively steep topography and complex land use.

• Proceedings of the Korean Association for Survey Research Conference
• /
• 2000.11a
• /
• pp.41-62
• /
• 2000

현실적으로 통계추론 방법의 적용시, 그 정당성이 보장되는 기본가정이외에도 추가적인 가정이 불가피하여, 본래의 정당성이 퇴색되는 경우가 흔히 발생한다. 따라서 이런 경우에는 통계추론의 평가가 필수적일 것이나, 많은 경우에 분석적 평가를 하기에는 너무 복잡하여, 특정상황을 상정한 모의분석 평가가 주류를 이루고 있다. 본 고에서는 보다 일반적 상황에서의 통계추론의 평가를 위해 브트스트랩방법과 같이 관찰값에 의존한 모의방법(observation-based simulation)을 이용한 평가방법을 제안한다. 우선 설득력 있는 평가요소로서 구간추정시 포함확률(coverage probability)와 같은 빈도성질(frequency property)를 선택하였다. 빈도성질은 고전적 통계추론은 물론 베이지안 통계추론을 대상으로도 의미있는 평가기준으로 판단되는 바, 이를 평가요소로서 선택하고, 이의 추정을 위한 방법과, 그 추정결과의 해석과 나아가 이를 기준으로 한 통계추론 결과의 조정 방법까지 일련의 절차에 대한 방법론을 제시하였다.