• Communications for Statistical Applications and Methods
• /
• 제24권1호
• /
• pp.67-80
• /
• 2017

Over the last few decades, ensemble forecasts based on global climate models have become an important part of climate forecast due to the ability to reduce uncertainty in prediction. Moreover in ensemble forecast, assessing the prediction uncertainty is as important as estimating the optimal weights, and this is achieved through a probabilistic forecast which is based on the predictive distribution of future climate. The Bayesian model averaging has received much attention as a tool of probabilistic forecasting due to its simplicity and superior prediction. In this paper, we propose a new Bayesian model averaging method for probabilistic ensemble forecasting. The proposed method combines a deterministic ensemble forecast based on a multivariate regression approach with Bayesian model averaging. We demonstrate that the proposed method is better in prediction than the standard Bayesian model averaging approach by analyzing monthly average precipitations and temperatures for ten cities in Korea.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권9호
• /
• pp.4684-4705
• /
• 2019

Collaborative filtering recommender systems are vulnerable to shilling attacks in which malicious users may inject biased profiles to promote or demote a particular item being recommended. To tackle this problem, many robust collaborative recommendation methods have been presented. Unfortunately, the robustness of most methods is improved at the expense of prediction accuracy. In this paper, we construct a robust Bayesian probabilistic matrix factorization model for collaborative filtering recommender systems by incorporating the detection of user anomaly rating behaviors. We first detect the anomaly rating behaviors of users by the modified K-means algorithm and target item identification method to generate an indicator matrix of attack users. Then we incorporate the indicator matrix of attack users to construct a robust Bayesian probabilistic matrix factorization model and based on which a robust collaborative recommendation algorithm is devised. The experimental results on the MovieLens and Netflix datasets show that our model can significantly improve the robustness and recommendation accuracy compared with three baseline methods.

• 한국지능시스템학회논문지
• /
• 제11권9호
• /
• pp.849-854
• /
• 2001

Bishop이 제안한 Generative Topographic Mapping(GTM)은 Kohonen이 제안한 자율 학습 신경망인 Self Organizing Maps(SOM)의 확률 버전이다. GTM은 데이터가 생성되는 확률 분포를 잠재 변수, 혹은 은닉 변수를 사용하여 모형화한다. 이것은 SOM에서는 구현될 수 없는 GTM만의 특징이며, 이러한 특징으로 인하여 SOM의 한계들을 극복할 수 있게 된다. 본 논문에서는 이러한 GTM 모형에 베이지안 학습(Bayesian learning)을 결합하여 작은 오분류율을 가지는 분류 알고리즘인 베이지안 GTM(Bayesian GTM)을 제안한다. 이 알고리즘은 기존의 GTM의 빠른 계산 처리 능력과 데이터에 대한 확률 분포, 그리고 베이지안 추론의 정확성을 이용하여 기존의 분류 알고리즘보다 우수한 결과를 얻게 된다. 본 논문에서는 기존의 분류 알고리즘에서 많이 실험하였다. 학습 데이터를 통하여 이를 확인하였다.

• 전자공학회논문지S
• /
• 제35S권4호
• /
• pp.105-117
• /
• 1998

In this paper, the general formula of disparity estimation based on Bayesian Maximum A Posteriori (MAP) algorithm is derived and implemented with simplified probabilistic models. The probabilistic models are independence and similarity among the neighboring disparities in the configuration.The formula is the generalized probabilistic diffusion equation based on Bayesian model, and can be implemented into the some different forms corresponding to the probabilistic models in the disparity neighborhood system or configuration. And, we proposed new probabilistic models in order to simplify the joint probability distribution of disparities in the configuration. According to the experimental results, the proposed algorithm outperformed the other ones, such as sum of swuared difference(SSD) based algorithm and Scharstein's method. We canconclude that the derived formular generalizes the probabilistic diffusion based on Bayesian MAP algorithm for disparity estimation, and the propsoed probabilistic models are reasonable and approximate the pure joint probability distribution very well with decreasing the computations to 0.01% of the generalized formula.

• Smart Structures and Systems
• /
• 제21권5호
• /
• pp.601-609
• /
• 2018

The estimated probabilistic model of wind data based on the conventional approach may have high discrepancy compared with the true distribution because of the uncertainty caused by the instrument error and limited monitoring data. A sequential quadratic programming (SQP) algorithm-based finite mixture modeling method has been developed in the companion paper and is conducted to formulate the joint probability density function (PDF) of wind speed and direction using the wind monitoring data of the investigated bridge. The established bivariate model of wind speed and direction only represents the features of available wind monitoring data. To characterize the stochastic properties of the wind parameters with the subsequent wind monitoring data, in this study, Bayesian inference approach considering the uncertainty is proposed to update the wind parameters in the bivariate probabilistic model. The slice sampling algorithm of Markov chain Monte Carlo (MCMC) method is applied to establish the multi-dimensional and complex posterior distribution which is analytically intractable. The numerical simulation examples for univariate and bivariate models are carried out to verify the effectiveness of the proposed method. In addition, the proposed Bayesian inference approach is used to update and optimize the parameters in the bivariate model using the wind monitoring data from the investigated bridge. The results indicate that the proposed Bayesian inference approach is feasible and can be employed to predict the bivariate distribution of wind speed and direction with limited monitoring data.

• 한국신뢰성학회지:신뢰성응용연구
• /
• 제18권3호
• /
• pp.271-279
• /
• 2018

Purpose: Probabilistic safety analysis was performed to enhance the safety and reliability of nuclear power plants because traditional deterministic approach has limitations in predicting the risk of failure by crack growth. The study introduces a probabilistic approach to establish a basis for probabilistic safety assessment of passive components. Methods: For probabilistic modeling of fatigue crack growth rate (FCGR), various FCGR tests were performed either under constant load amplitude or constant ${\Delta}K$ conditions by using heat treated X-750 at low temperature with adequate cathodic polarization. Bayesian inference was employed to update uncertainties of the FCGR model using additional information obtained from constant ${\Delta}K$ tests. Results: Four steps of Bayesian parameter updating were performed using constant ${\Delta}K$ test results. The standard deviation of the final posterior distribution was decreased by a factor of 10 comparing with that of the prior distribution. Conclusion: The method for developing a probabilistic crack growth model has been designed and demonstrated, in the paper. Alloy X-750 has been used for corrosion fatigue crack growth experiments and modeling. The uncertainties of parameters in the FCGR model were successfully reduced using the Bayesian inference whenever the updating was performed.

• Nuclear Engineering and Technology
• /
• 제53권1호
• /
• pp.304-313
• /
• 2021

Nickel base Alloy X-750, which is used as fastener parts in light-water reactor (LWR), has experienced many failures by environmentally assisted cracking (EAC). In order to improve the reliability of passive components for nuclear power plants (NPP's), it is necessary to study the failure mechanism and to predict crack growth behavior by developing a probabilistic failure model. In this study, The Bayesian inference was employed to reduce the uncertainties contained in EAC modeling parameters that have been established from experiments with Alloy X-750. Corrosion fatigue crack growth rate model (FCGR) was developed by fitting into Paris' Law of measured data from the several fatigue tests conducted either in constant load or constant ΔK mode. These parameters characterizing the corrosion fatigue crack growth behavior of X-750 were successfully updated to reduce the uncertainty in the model by using the Bayesian inference method. It is demonstrated that probabilistic failure models for passive components can be developed by updating a laboratory model with field-inspection data, when crack growth rates (CGRs) are low and multiple inspections can be made prior to the component failure.

• Advances in concrete construction
• /
• 제9권4호
• /
• pp.337-343
• /
• 2020

In the present study, by employing fragility analysis, the seismic vulnerability of a concrete girder bridge, one of the most common existing structural bridge systems, has been performed. To this end, drift demand model as a fundamental ingredient of any probabilistic decision-making analyses is initially developed in terms of the two most common intensity measures, i.e., PGA and Sa (T₁). Developing a probabilistic demand model requires a reliable database that is established in this paper by performing incremental dynamic analysis (IDA) under a set of 20 ground motion records. Next, by employing Bayesian statistical inference drift demand models are developed based on pre-collapse data obtained from IDA. Then, the accuracy and reasonability of the developed models are investigated by plotting diagnosis graphs. This graphical analysis demonstrates probabilistic demand model developed in terms of PGA is more reliable. Afterward, fragility curves according to PGA based-demand model are developed.

• 국제학술발표논문집
• /
• The 5th International Conference on Construction Engineering and Project Management
• /
• pp.488-493
• /
• 2013

Accurate initial cost estimates are essential to effective management of construction projects where many decisions are made in the course of project management by referencing the estimates. In practice, the initial estimates are frequently derived from historical actual cost data, for which standard distribution-based techniques are widely applied in the construction industry to account for risk associated with the estimates. This approach assumes the same probability distribution of estimate errors for any selected estimates. This assumption, however, is not always satisfied. In order to account for the probabilistic nature of estimate errors, an alternative method for measuring the risk associated with a selected initial estimate is developed by applying the Bayesian probability approach. An application example include demonstrates how the method is implemented. A hypothesis test is conducted to reveal the robustness of the Bayesian probability model. The method is envisioned to effectively complement cost estimating methods that are currently in use by providing benefits as follows: (1) it effectively accounts for the probabilistic nature of errors in estimates; (2) it is easy to implement by using historical estimates and actual costs that are readily available in most construction companies; and (3) it minimizes subjective judgment by using quantitative data only.

• 한국산업정보학회논문지
• /
• 제23권6호
• /
• pp.79-94
• /
• 2018

마케팅 분야에서는 제품품질, 고객만족, 고객추천을 바탕으로 구매행동과의 영향 유무 및 상관관계를 통계적 Regression 방법으로 가설 검증하는 것을 주요한 연구 대상으로 하고 있다. 또한 최근에는 ASCI와 같은 고객만족지수 혹은 라이켈트의 NPS와 같은 고객추천지수를 바탕으로 실제 기업성과와 연관되는 시장 지분에 어떠한 영향을 미치는 지에 대한 통계적 분석 연구도 활발히 이루어지고 있다. 본 연구에서는 실제 고객이 매장을 방문하여, 과거 고객카드에 명품을 구매하던 구매하지 않던 간에 만족/불만족을 표시한 체인 및 고객 추천의향을 검토하여 Hidden Markov Model을 이용한 고객의 최상의 구매패턴을 분석하는 확률적 기법에 대하여 연구하는 것을 목적으로 하고 있다. 이를 바탕으로 고객만족 -> 고객추천의향 -> 고객추천행동->구매 및 재구매 체인에 대응하는 실제 소비자의 구매패턴을 고객만족과 NPS(순추천지수) 및 여러 수리통계적 이론-Hidden Markov Model, Bayesian Inference, Maximum Likelihood Estimation을 이용하여 확률적 추적 메카니즘을 구현하는 것을 목표로 한다. 제시된 목표는 인공지능을 구현하는 이론과 알고리듬을 사용하여 달성되었기에 이론적 추적 메카니즘을 여러 인공지능망 -DNN, CNN, GAN등을 사용하여 기업에서 사용할 수 있는 고객의 구매패턴 앱으로 발전시키는 것을 후속연구에서 기대한다.