• 정보처리학회논문지D
• /
• 제13D권3호
• /
• pp.369-376
• /
• 2006

본 논문은 온라인 전자문서환경에서 전통적 베이지안 통계기반 문서분류시스템의 분류성능을 개선하기 위해 EM(Expectation Maximization) 가속 알고리즘을 접목한 방법을 제안한다. 기계학습 기반의 문서분류시스템의 중요한 문제 중의 하나는 양질의 학습문서를 확보하는 것이다. EM 알고리즘은 소량의 학습문서집합으로 베이지안 문서분류 알고리즘의 성능을 높이는데 활용된다. 그러나 EM 알고리즘은 최적화 과정에서 느린 수렴성과 성능 저하 현상을 나타내는데, EM 알고리즘의 기본 가정을 따르지 않는 온라인 전자문서환경에서 특히 그러하다. 제안 기법의 주요 아이디어는 전통적 EM 알고리즘을 개선하기 위해 불확정성도 기반 선택적 샘플링 기법을 활용한 것이다. 성능평가를 위해 Reuter-21578 문서집합을 사용하여, 제안 알고리즘의 빠른 수렴성을 보이고 전통적 베이지안 알고리즘의 분류 정확성을 향상시켰음을 보인다.

• 한국컴퓨터산업학회논문지
• /
• 제4권12호
• /
• pp.1043-1052
• /
• 2003

본 논문에서는 비디오코딩의 움직임 추정을 위한 빠른 움직임 추정알고리즘을 제안하였다. 제안된 알고리즘은 다단계 연속제거 알고리즘과 효율적인 다단계 연속제거 알고리즘에 기초하고 있다. 제안된 알고리즘은 계층적으로 탐색점을 추출하여 매우 많은 연산량을 필요로 하는 정합 연산량을 감소시키면서 최상의 움직임벡터를 얻을 수 있다. 실험을 통하여 제안된 알고리즘의 효율성을 확인하였다.

• Journal of the Korean Statistical Society
• /
• 제23권2호
• /
• pp.463-482
• /
• 1994

In Bayesian analysis of randomized response models, the likelihood function does not combine tractably with typical priors for the parameters of interest, causing computational difficulties in posterior analysis of the parameters of interest. In this article, the difficulties are solved by introducing appropriate latent variables to the model and using the Gibbs sampling algorithm.

• Communications for Statistical Applications and Methods
• /
• 제7권1호
• /
• pp.211-224
• /
• 2000

The Bayes estimation of the proportion in successive occasions sampling with randomized response model is discussed by means of Acceptance Rejection sampling. Bayesian estimation of transition probabilities in two successive occasions is suggested via Markov Chain Monte Carlo algorithm and its applicability is represented in a numerical example.

• 한국CDE학회논문집
• /
• 제15권1호
• /
• pp.24-32
• /
• 2010

This paper introduces a B-spline based branch & bound algorithm for global optimization. The branch & bound is a well-known algorithm paradigm for global optimization, of which key components are the subdivision scheme and the bound calculation scheme. For this, we consider the B-spline hypervolume to approximate an objective function defined in a design space. This model enables us to subdivide the design space, and to compute the upper & lower bound of each subspace where the bound calculation is based on the LHS sampling points. We also describe a search tree to represent the searching process for optimal solution, and explain iteration steps and some conditions necessary to carry out the algorithm. Finally, the performance of the proposed algorithm is examined on some test problems which would cover most difficulties faced in global optimization area. It shows that the proposed algorithm is complete algorithm not using heuristics, provides an approximate global solution within prescribed tolerances, and has the good possibility for large scale NP-hard optimization.

• 대한전기학회논문지:전기기기및에너지변환시스템부문B
• /
• 제53권3호
• /
• pp.206-211
• /
• 2004

In phase controlled rectifier, it's been known that a fast response is achieved by predictive current control without any overshoot. The frequent sampling period is essential to improve the firing accuracy in conventional predict current control. However, improving the firing accuracy if difficult to reduce the period of sampling efficiently because current sampling and predictive current control is carried out in every period and the ON-OFF current control is performed by comparing two different one. To improve the firing accuracy at the predictive current control, the calculated firing angle is loaded into the high-accuracy hardware timer. So the calculation of exact crossing point between the predictive and actual current is the most important. In this paper, the flow chart for proposed firing angle calculation algorithm is obtained for the fastest current control performance in transient state. The performance of proposed algorithm is verified through simulations and experiments.

• 조명전기설비학회논문지
• /
• 제24권12호
• /
• pp.138-146
• /
• 2010

On wind turbine generators, the speed and volume of the wind affect the turbine angle speed which finally determines the output level of the electric power. However it is very difficult to forecast correctly the future power output and quality based on previous fixed sampling methods. This paper proposes a variable sampling method based on Buneman frequency estimation algorithm to reflect the variations of the frequency and amplitude on wind power outputs. The proposed method is also verified through the performance test by comparing with the results from previous fixed sampling methods and the real measurement data.

• ETRI Journal
• /
• 제14권4호
• /
• pp.19-27
• /
• 1992

Since. as is well known, direct computation of the reliability for a large-scaled and complex net work generally requires exponential time, a variety of alternative methods to estimate the network reliability using simulation have been proposed. Monte Carlo sampling is the major approach to estimate the network reliability using simulation. In the paper, a dynamic Monte Carlo sampling method, called conditional minimal cut set (CMCS) algorithm, is suggested. The CMCS algorithm simulates a minimal cut set composed of arcs originated from the (conditional) source node until s-t connectedness is confirmed, then reduces the network on the basis of the states of simulated arcs. We develop the importance sampling estimator and the total hazard estimator and compare the performance of these simulation estimators. It is found that the CMCS algorithm is useful in reducing variance of network reliability estimator.

• 한국데이터정보과학회:학술대회논문집
• /
• 한국데이터정보과학회 2004년도 춘계학술대회
• /
• pp.39-50
• /
• 2004

The decision tree approach is most useful in classification problems and to divide the search space into rectangular regions. Decision tree algorithms are used extensively for data mining in many domains such as retail target marketing, fraud dection, data reduction and variable screening, category merging, etc. CHAID(Chi-square Automatic Interaction Detector) uses the chi-squired statistic to determine splitting and is an exploratory method used to study the relationship between a dependent variable and a series of predictor variables. In this paper we propose CHAID algorithm by cube-based proportional sampling and explore CHAID algorithm in view of accuracy and speed by the number of variables.

• Journal of the Korean Data and Information Science Society
• /
• 제15권4호
• /
• pp.803-816
• /
• 2004

The decision tree approach is most useful in classification problems and to divide the search space into rectangular regions. Decision tree algorithms are used extensively for data mining in many domains such as retail target marketing, fraud dection, data reduction and variable screening, category merging, etc. CHAID uses the chi-squired statistic to determine splitting and is an exploratory method used to study the relationship between a dependent variable and a series of predictor variables. In this paper we propose CHAID algorithm by cube-based proportional sampling and explore CHAID algorithm in view of accuracy and speed by the number of variables.