• Communications for Statistical Applications and Methods
• /
• 제10권2호
• /
• pp.595-601
• /
• 2003

In this paper we propose an on line training method for classification based on least squares support vector machine. Proposed method enables the computation cost to be reduced and the training to be peformed incrementally, With the incremental formulation of an inverse matrix in optimization problem, current information and new input data can be used for building the new inverse matrix for the estimation of the optimal bias and Lagrange multipliers, so the large scale matrix inversion operation can be avoided. Numerical examples are included which indicate the performance of proposed algorithm.

• Journal of the Korean Data and Information Science Society
• /
• 제25권4호
• /
• pp.951-959
• /
• 2014

Recently Hazelton and Turlach (2009) proposed a weighted kernel density estimator for the deconvolution problem. In the case of Gaussian kernels and measurement error, they argued that the weighted kernel density estimator is a competitive estimator over the classical deconvolution kernel estimator. In this paper we consider weighted kernel density estimators when sample observations are contaminated by double exponentially distributed errors. The performance of the weighted kernel density estimators is compared over the classical deconvolution kernel estimator and the kernel density estimator based on the support vector regression method by means of a simulation study. The weighted density estimator with the Gaussian kernel shows numerical instability in practical implementation of optimization function. However the weighted density estimates with the double exponential kernel has very similar patterns to the classical kernel density estimates in the simulations, but the shape is less satisfactory than the classical kernel density estimator with the Gaussian kernel.

• 산업기술연구
• /
• 제19권
• /
• pp.209-216
• /
• 1999

In this paper, we proposed a modofied genetic algorithm to minimize misclassification rate for determining the codebook. Genetic algorithms are adaptive methods which may be used solve search and optimization problems based on the genetic processes of biological organisms. But they generally require a large amount of computation efforts. GAVQ can choose the optimal individuals by genetic operators. The position of individuals are optimized to improve the recognition rate. The technical properties of this study is that prevents us from the local minimum problem, which is not avoidable by conventional VQ algorithms. We compared the simulation result with Matlab using phoneme data. The simulation results show that the recognition rate from GAVQ is improved by comparing the conventional VQ algorithms.

• 응용통계연구
• /
• 제18권2호
• /
• pp.421-434
• /
• 2005

기계학습은 패턴분류의 한 도구로써 광범위하게 연구되고 있다. 기계학습 방법들 중에서 서포트 벡터 기계(Support Vector Machines)는 많은 분야에서 연구되어지는 것으로 이진 패턴 분류문제에서 고차원의 특징공간에서 두 집합들 사이에 가장 큰 분리를 제공하는 최대 여유도(margin)를 가지는 분리 초평면을 찾는 것이다. 최대 여유도의 분리의 개념에 기초하여 Mangasarian(1968)은 다중-표면 방법(multi-surface method)을 제안하였고, 1980년대에 목적 계획법을 이용한 방법들이 광범위하게 개발되었다. 본 논문에서는 다목적 계획법과 목적 계획법을 이용한 수리계획법인 서포트 벡터 기계의 두가지 방법들을 제안하고 수치 예제들을 통하여 효용성에 대하여 논의하고자 한다.

• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 2005년도 춘계 학술발표회 논문집
• /
• pp.335-342
• /
• 2005

Using higher order Mindlin plates and piezoelectric materials, eigenvalue problems are considered. Since piezoelectric crystal resonators produce a proper amount of electric signal for a thickness-shear frequency, the objective is to decouple the thickness-shear mode from the others. Design variables are the bulk material densities corresponding to the mass of masking plates for electrodes. The design sensitivity expressions for the thickness-shear frequency and mode shape vector are derived using direct differentiation method(DDM). Using the developed design sensitivity analysis (DSA) method, we formulate a topology optimization problem whose objective function is to maximize the thickness-shear component of strain energy density at the thickness-shear mode. Constraints are the allowable volume and area of masking plate. Numerical examples show that the optimal design yields an improved mode shape and thickness-shear energy.

• 한국생산제조학회지
• /
• 제20권4호
• /
• pp.412-418
• /
• 2011

Reliability-based Topology optimization(RBTO) is to get an optimal design satisfying uncertainties of design variables. Although RBTO based on homogenization and density distribution method has been done, RBTO based on BESO has not been reported yet. This study presents a reliability-based topology optimization(RBTO) using bi-directional evolutionary structural optimization(BESO). Topology optimization is formulated as volume minimization problem with probabilistic displacement constraint. Young's modulus, external load and thickness are considered as uncertain variables. In order to compute reliability index, four methods, i.e., RIA, PMA, SLSV and ADL(adaptive-loop), are used. Reliability-based topology optimization design process is conducted to obtain optimal topology satisfying allowable displacement and target reliability index with the above four methods, and then each result is compared with respect to numerical stability and computing time. The results of this study show that the RBTO based on BESO using the four methods can effectively be applied for topology optimization. And it was confirmed that DLSV and ADL had better numerical efficiency than SLSV. ADL and SLSV had better time cost than DLSV. Consequently, ADL method showed the best time efficiency and good numerical stability.

• 한국지능정보시스템학회:학술대회논문집
• /
• 한국지능정보시스템학회 2005년도 공동추계학술대회
• /
• pp.516-525
• /
• 2005

Case-based reasoning (CBR) has been widely used in various areas due to its convenience and strength in complex problem solving. Generally, in order to obtain successful results from CBR, effective retrieval of useful prior cases for the given problem is essential. However, designing a good matching and retrieval mechanism for CBR systems is still a controversial research issue. Most prior studies have tried to optimize the weights of the features or selection process of appropriate instances. But, these approaches have been performed independently until now. Simultaneous optimization of these components may lead to better performance than in naive models. In particular, there have been few attempts to simultaneously optimize the weight of the features and selection of the instances for CBR. Here we suggest a simultaneous optimization model of these components using a genetic algorithm (GA). We apply it to a customer classification model which utilizes demographic characteristics of customers as inputs to predict their buying behavior for a specific product. Experimental results show that simultaneously optimized CBR may improve the classification accuracy and outperform various optimized models of CBR as well as other classification models including logistic regression, multiple discriminant analysis, artificial neural networks and support vector machines.

• 산업경영시스템학회지
• /
• 제43권4호
• /
• pp.208-216
• /
• 2020

With the spread of smart manufacturing, one of the key topics of the 4th industrial revolution, manufacturing systems are moving beyond automation to smartization using artificial intelligence. In particular, in the existing automatic machining, a number of machining defects and non-processing occur due to tool damage or severe wear, resulting in a decrease in productivity and an increase in quality defect rates. Therefore, it is important to measure and predict tool life. In this paper, ν-ASVR (ν-Asymmetric Support Vector Regression), which considers the asymmetry of ⲉ-tube and the asymmetry of penalties for data out of ⲉ-tube, was proposed and applied to the tool wear prediction problem. In the case of tool wear, if the predicted value of the tool wear amount is smaller than the actual value (under-estimation), product failure may occur due to tool damage or wear. Therefore, it can be said that ν-ASVR is suitable because it is necessary to overestimate. It is shown that even when adjusting the asymmetry of ⲉ-tube and the asymmetry of penalties for data out of ⲉ-tube, the ratio of the number of data belonging to ⲉ-tube can be adjusted with ν. Experiments are performed to compare the accuracy of various kernel functions such as linear, polynomial. RBF (radialbasis function), sigmoid, The best result isthe use of the RBF kernel in all cases

• 융합신호처리학회논문지
• /
• 제10권1호
• /
• pp.32-39
• /
• 2009

동화상의 부호화 효율향상을 위하여 유전적 알고리즘(Genetic Algorithm: GA)을 이용하여 영역 분할하는 방법을 제안한다. 유전적 알고리즘은 함수치만을 이용하여 큰 탐색공간으로부터 최적의 조합을 축차적으로 찾아내는 방법이다. 이동추정과 영역분할을 동시에 진행함으로써, 이동 벡터를 화면내의 작은 블록이나 화소의 각각에 할당하고, 그것을 부호화 정보량과 신호 대 잡음비의 관계로부터 최적화 문제로 변환할 수 있다. 즉, 이동보상예측 부호화에는 영역분할과 이동 추정은 서로 밀접하게 관계되어 있다. 이것은 부호량과 S/N비를 최적화하는 것으로서 화면 속의 각 블록에서 이동 벡타를 최적의 상태로 배치하는 것이다. 그러므로, 본 논문에서는 최적인 영역분할 결과를 얻기 위하여 GA의 데이터형과 그 데이터의 처리 방법에 대해서 검토하였다. 또한, 테스트 화상을 이용한 컴퓨터시뮬레이션을 통하여 제안 방법의 유효성을 확인하였다.

• 한국경영과학회지
• /
• 제18권3호
• /
• pp.13-34
• /
• 1993

By operating on many part of a software system concurrently, the parallel processing computers may provide several orders of magnitude more computing power than traditional serial computers. If the Lagrangean approximation procedure is applied to a large scale manufacturing problem which is decomposable into many subproblems, the procedure is a perfect candidate for parallel processing. By distributing Lagrangean subproblems for given multiplier to multiple processors, concurrently running processors and modifying Lagrangean multipliers at the end of each iteration of a subgradient method,a parallel processing of a Lagrangean approximation procedure may provide a significant speedup. This purpose of this research is to investigate the potential of the parallelized Lagrangean approximation procedure (PLAP) for certain combinational optimization problems in manufacturing systems. The framework of a Plap is proposed for some combinatorial manufacturing problems which are decomposable into well-structured subproblems. The synchronous PLAP for the multistage dynamic lot-sizing problem is implemented on a parallel computer Alliant FX/4 and its computational experience is reported as a promising application of vector-concurrent computing.