• Journal of KIISE:Software and Applications
• /
• v.28 no.9
• /
• pp.681-687
• /
• 2001

In order to raise a class discrimination power by combining multiple classifiers under the Bayesian decision theory, the upper bound of a Bayes error rate bounded by the conditional entropy of a class variable and decision variables obtained from training data samples should be minimized. Wang and Wong proposed a tree dependence first-order approximation scheme of a high order probability distribution composed of the class and multiple feature pattern variables for minimizing the upper bound of the Bayes error rate. This paper presents an extended high order product approximation scheme dealing with higher order dependency more than the first-order tree dependence, based on the minimization of the upper bound of the Bayes error rate. Multiple recognizers for unconstrained handwritten numerals from CENPARMI were combined by the proposed approximation scheme using the Bayesian formalism, and the high recognition rates were obtained by them.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2017.05a
• /
• pp.87-87
• /
• 2017

최근 이상기후로 인해 국내 가뭄피해가 증가하고 있는 추세이며, 미래 가뭄의 심도 및 지속시간은 증가할 것으로 예측되고 있다. 특히 우리나라는 용수공급의 56.5%를 댐에 의존하여 댐 유역의 가뭄은 생 공 농업용수 공급제한 등의 광범위한 피해를 발생시킬 수 있다. 다만 가뭄은 홍수와 달리 진행속도가 비교적 느리기 때문에 사전에 정확한 댐 유입량 예측이 가능하다면, 용수공급량 조정을 통해 피해를 최소화할 수 있다. 국내에서는 댐 유입량 예측에 ESP (Ensemble Streamflow Prediction) 기법을 활용하고 있으며, ESP 기법은 과거 기상자료를 기반으로 미래를 예측하기 때문에 기상자료, 초기수문조건, 매개변수 등에 불확실성을 가지고 있다. 본 연구에서는 베이지안 이론을 이용하여 댐 예측유입량의 정확도 향상기법을 개발하고 예측성을 평가하고자 하며, 강우유출모델은 ABCD를 활용하였다. 대상유역은 국내의 대표 다목적댐인 충주댐 유역을 선정하였으며, 기상자료는 기상청, 국토교통부 및 한국수자원공사의 지점자료를 수집하였다. 예측성 평가기법으로는 도시적 분석방법인 시계열 분석, 통계적 분석방법인 Skill Score (SS)를 활용하였다. 시계열 분석 결과 ESP 댐 예측유입량(ESP)은 매년 월별 전망값의 큰 차이가 없었으며, 다우년 및 과우년의 예측성이 떨어지는 것으로 나타났다. 베이지안 기반의 댐 예측유입량(BAYES-ESP)는 ESP의 과소모의하는 경향을 보정하였으며, 다우년에 예측성이 향상되었다. 월별 평균 댐 관측유입량과 ESP, BAYES-ESP의 SS 비교분석 결과 ESP는 유입량 값이 적은 1, 2, 3월에 SS가 양의 값을 가졌으며, 이외의 월에는 음의 값으로 나타났다. BAYES-ESP는 ESP와 관측값이 비교적 선형관계를 나타내는 1, 2, 3월에 ESP의 예측성을 개선시키는 것으로 나타났다. ESP 기법은 강수량의 월별, 계절별 변동성이 큰 우리나라에 적용하기에는 예측성의 한계가 있었으며, 이를 개선한 BAYES-ESP 기법은 댐 유입량 예측 연구에 가치가 있는 것으로 판단된다.

• The Korean Journal of Applied Statistics
• /
• v.27 no.1
• /
• pp.71-87
• /
• 2014

The Bayes premium or the net premium in the credibility theory does not reflect the underlying tail risk. In this study we examine how the tail risk measures can be utilized in determining the risk premium. First, we show that the risk measures can not only provide the proper risk loading, but also allow the insurer to avoid the wrong decision made with the Bayesian premium alone. Second, it is illustrated that the rank of the tail thickness among different conditional loss distributions does not preserve for the corresponding predictive distributions, even if they share the identical prior variable. The implication of this result is that the risk loading for a contract should be based on the risk measure of the predictive loss distribution not the conditional one.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.38 no.6
• /
• pp.793-800
• /
• 2018

It is necessary to determine an optimal design frequency for establishing stable flood control against frequent flood disasters. Depending on the importance of river and regional characteristics, design return periods are suggested from at least 50 years up to 200 years for river design. However, due to the wide range of applications, it is not desirable to reflect the geographical and flood control characteristics of river. In this study, Bayes theory was applied to seven evaluation factors to determine the optimal design return period of rivers in Chungcheongnam-do; urbanization flooded area, watershed area, basin coefficient, slope, water system and stream order, range of backwater effect, abnormal rainfall occurrence frequency. The potential flood damage (PFD) capacity was estimated considering climate change and the appropriate design return period was determined by analyzing the capacity of each district. We compared the design return periods of 382 rivers in Chungcheongnam-do with the existing design return periods. The number of rivers that were upgraded from the existing return period were 65, which have relatively large flooding areas and have large PFDs. Whereas, the number of rivers that were downgraded were 169.

• Journal of the Korean Statistical Society
• /
• v.36 no.3
• /
• pp.367-372
• /
• 2007

Classification as a tool to extract information from data plays an important role in science and engineering. Among various classification methodologies, support vector machine has recently seen significant developments. The central problem this paper addresses is the accuracy of support vector machine. In particular, we are interested in the situations where fast rates of convergence to the Bayes risk can be achieved by support vector machine. Through learning examples, we illustrate that support vector machine may yield fast rates if the space spanned by an adopted kernel is sufficiently large.

• Proceedings of the Korean Information Science Society Conference
• /
• 2012.06b
• /
• pp.522-524
• /
• 2012

For exact classification of the defect, good feature selection and classifier is necessary. In this paper, various features such as brightness features, shape features and statistical features are stated and Bayes classifier using Gaussian mixture model is used as classifier. Also feature extraction method based on spectral graph theory is presented. Experimental result shows that feature extraction method using graph Laplacian result in better performance than the result using PCA.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.6
• /
• pp.2470-2491
• /
• 2018

Web-scale open information extraction (Open IE) plays an important role in NLP tasks like acquiring common-sense knowledge, learning selectional preferences and automatic text understanding. A large number of Open IE approaches have been proposed in the last decade, and the majority of these approaches are based on supervised learning or dependency parsing. In this paper, we present a novel method for web scale open information extraction, which employs cosine distance based on Google word vector as the confidence score of the extraction. The proposed method is a purely unsupervised learning algorithm without requiring any hand-labeled training data or dependency parse features. We also present the mathematically rigorous proof for the new method with Bayes Inference and Artificial Neural Network theory. It turns out that the proposed algorithm is equivalent to Maximum Likelihood Estimation of the joint probability distribution over the elements of the candidate extraction. The proof itself also theoretically suggests a typical usage of word vector for other NLP tasks. Experiments show that the distance-based method leads to further improvements over the newly presented Open IE systems on three benchmark datasets, in terms of effectiveness and efficiency.

• Journal of Information Technology Services
• /
• v.11 no.1
• /
• pp.211-221
• /
• 2012

Many metric-based classification models have been proposed to predict fault-proneness of software module. This paper presents two prediction models using Bayesian classifier which is one of the most popular modern classification algorithms. Bayesian model based on Bayesian probability theory can be a promising technique for software quality prediction. This is due to the ability to represent uncertainty using probabilities and the ability to partly incorporate expert's knowledge into training data. The two models, Na$\ddot{i}$veBayes(NB) and Bayesian Belief Network(BBN), are constructed and dimensionality reduction of training data and test data are performed before model evaluation. Prediction accuracy of the model is evaluated using two prediction error measures, Type I error and Type II error, and compared with well-known prediction models, backpropagation neural network model and support vector machine model. The results show that the prediction performance of BBN model is slightly better than that of NB. For the data set with ambiguity, although the BBN model's prediction accuracy is not as good as the compared models, it achieves better performance than the compared models for the data set without ambiguity.

• International Journal of Computer Science & Network Security
• /
• v.23 no.9
• /
• pp.192-198
• /
• 2023

Cyberbullying is a growing problem among adolescents and can have serious psychological and emotional consequences for the victims. In recent years, machine learning techniques have emerged as promising approach for detecting instances of cyberbullying in online communication. This research paper focuses on developing a machine learning models that are able to detect cyberbullying including support vector machines, naïve bayes, and random forests. The study uses a dataset of real-world examples of cyberbullying collected from Twitter and extracts features that represents the ideational metafunction, then evaluates the performance of each algorithm before and after considering the theory of systemic functional linguistics in terms of precision, recall, and F1-score. The result indicates that all three algorithms are effective at detecting cyberbullying with 92% for naïve bayes and an accuracy of 93% for both SVM and random forests. However, the study also highlights the challenges of accurately detecting cyberbullying, particularly given the nuanced and context-dependent nature of online communication. This paper concludes by discussing the implications of these findings for future research and the development of practical tool for cyberbullying prevention and intervention.

• Journal of Communications and Networks
• /
• v.17 no.1
• /
• pp.75-83
• /
• 2015

In mobile ad-hoc networks (MANETs), nodes are mobile in nature. Collaboration between mobile nodes is more significant in MANETs, which have as their greatest challenges vulnerabilities to various security attacks and an inability to operate securely while preserving its resources and performing secure routing among nodes. Therefore, it is essential to develop an effective secure routing protocol to protect the nodes from anonymous behaviors. Currently, game theory is a tool that analyzes, formulates and solves selfishness issues. It is seldom applied to detect malicious behavior in networks. It deals, instead, with the strategic and rational behavior of each node. In our study,we used the dynamic Bayesian signaling game to analyze the strategy profile for regular and malicious nodes. This game also revealed the best actions of individual strategies for each node. Perfect Bayesian equilibrium (PBE) provides a prominent solution for signaling games to solve incomplete information by combining strategies and payoff of players that constitute equilibrium. Using PBE strategies of nodes are private information of regular and malicious nodes. Regular nodes should be cooperative during routing and update their payoff, while malicious nodes take sophisticated risks by evaluating their risk of being identified to decide when to decline. This approach minimizes the utility of malicious nodes and it motivates better cooperation between nodes by using the reputation system. Regular nodes monitor continuously to evaluate their neighbors using belief updating systems of the Bayes rule.