• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2007년도 가을 학술발표논문집 Vol.34 No.2 (C)
• /
• pp.485-489
• /
• 2007

Gaussian process classifiers (GPCs) are fully statistical kernel classification models which have a latent function with Gaussian process prior Recently, EP approximation method has been proposed to infer the posterior over the latent function. It can have a special hyperparameter which can treat outliers potentially. In this paper, we propose the outlier robust algorithm which alternates EP and the hyperparameter updating until convergence. We also show its usefulness with the simulation results.

• 한국의학물리학회지:의학물리
• /
• 제15권2호
• /
• pp.100-104
• /
• 2004

선량 측정기의 공간적인 반응특성 때문에 나타나는 detector의 크기효과는 임상적인 선량측정을 부정확하게 만드는 중요한 원인이기에 많은 연구의 대상이 되어왔다. 관례적으로 detector response kernel은 detector 자체의 크기가 측정한 방사선의 선량분포에 대해 미친 영향에 대한 정보를 포함하고 있다. 이 kernel에 대해 다양한 수학적 모델들이 제안되었고 실험적으로 이론적으로 연구되어왔다. 이 논문은 convolution이론과 Monte Carlo simulation만을 이용하여 detector의 kernel을 결정하는 방법을 제시한다. 이 수치해석적인 방법을 사용하여 물 phantom에 잠긴 Farmer형 ion chamber의 detector response kernel을 계산하였다. 계산된 kernel은 기존의 parabolic 모델의 특성과 Gaussian 모델의 특성을 동시에 나타내고 있다. 이 kernel과 deconvolution 방법을 사용하여 측정된 6MV, 10${\times}$10 $\textrm{cm}^2$, 0.5${\times}$10 $\textrm{cm}^2$ 광자선으로부터 크기효과를 제거하였다. 크기효과가 제거된 방사선의 선량분포는 꼬리부분을 제외하고는 film이나 pin-point ion chamber에 의해 측정된 결과와 유사한 선량분포를 나타냈다.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2022년도 춘계학술대회
• /
• pp.144-146
• /
• 2022

본 논문에서는 Kernel based Fuzzy C-Means(K-FCM) 기반 양자화 기법을 적용하여 의료 초음파 영상에서 특징을 분할하는 기법을 제안한다. 결절종의 경우에는 초음파 영상 내에서 무에코, 저에코의 특징을 가진 낭포성 종양 객체를 특징 영역으로 영상을 분할한다. K-FCM 클러스터링은 기존의 FCM 클러스터링에서 Kernel Function을 적용한 형태의 클러스터링 기법이다. 본 논문에서는 Gaussian Kernel 기반 K-FCM을 적용하여 의료 초음파 영상에서 특징들을 분할하였다. 결절종 초음파 영상에서는 FCM 클러스터링이 F1 Score가 85.574%로 나타났고, K-FCM이 86.442%로 나타났다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권7호
• /
• pp.3076-3092
• /
• 2020

An IKPCA-ELM-based intrusion detection method is developed to address the problem of the low accuracy and slow speed of intrusion detection caused by redundancies and high dimensions of data in the network. First, in order to reduce the effects of uneven sample distribution and sample attribute differences on the extraction of KPCA features, the sample attribute mean and mean square error are introduced into the Gaussian radial basis function and polynomial kernel function respectively, and the two improved kernel functions are combined to construct a hybrid kernel function. Second, an improved particle swarm optimization (IPSO) algorithm is proposed to determine the optimal hybrid kernel function for improved kernel principal component analysis (IKPCA). Finally, IKPCA is conducted to complete feature extraction, and an extreme learning machine (ELM) is applied to classify common attack type detection. The experimental results demonstrate the effectiveness of the constructed hybrid kernel function. Compared with other intrusion detection methods, IKPCA-ELM not only ensures high accuracy rates, but also reduces the detection time and false alarm rate, especially reducing the false alarm rate of small sample attacks.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권4호
• /
• pp.2078-2093
• /
• 2019

Visual smoke recognition is a challenging task due to large variations in shape, texture and color of smoke. To improve performance, we propose a novel smoke recognition method by combining dual-encoded features that are extracted from both spatial and Curvelet domains. A Curvelet transform is used to filter an image to generate fifty sub-images of Curvelet coefficients. Then we extract Local Binary Pattern (LBP) maps from these coefficient maps and aggregate histograms of these LBP maps to produce a histogram map. Afterwards, we encode the histogram map again to generate Dual-encoded Local Binary Patterns (Dual-LBP). Histograms of Dual-LBPs from Curvelet domain and Completed Local Binary Patterns (CLBP) from spatial domain are concatenated to form the feature for smoke recognition. Finally, we adopt Gaussian Kernel Optimization (GKO) algorithm to search the optimal kernel parameters of Support Vector Machine (SVM) for further improvement of classification accuracy. Experimental results demonstrate that our method can extract effective and reasonable features of smoke images, and achieve good classification accuracy.

• 한국지능시스템학회논문지
• /
• 제12권4호
• /
• pp.300-305
• /
• 2002

본 논문에서는 타원형 클러스터링을 위한 거리측정 함수로써 변형된 가우시안 커널 함수를 사용하며, 주어진 클러스터링 문제를 각 타원형 클러스터의 체적을 최소화하는 문 로 해석하고 이를 선형행렬 부등식 기법 중 하나인 고유값 문제로 변환하여 최적화하는 새로운 알고리즘을 제안한다.

• 한국통신학회논문지
• /
• 제36권12C호
• /
• pp.753-758
• /
• 2011

The constant modulus algorithm (CMA) widely used in blind equalization applications minimizes the averaged power of constant modulus error (CME) defined as the difference between an instant output power and a constant modulus. In this paper, a decision feedback version of the linear blind algorithm based on maximization of the zero-error probability for CME is proposed. The Gaussian kernel of the maximum zero-error criterion is analyzed to have the property to cut out excessive CMEs that may be induced from severely distorted channel characteristics. Decision feedback approach to the maximum zero-error criterion for CME is developed based on the characteristic that the Gaussian kernel suppresses the outliers and this prevents error propagation to some extent. Compared to the linear algorithm based on maximum zero-error probability for CME in the simulation of blind equalization environments, the proposed decision feedback version has superior performance enhancement particularly in cases of severe channel distortions.

• 한국수학사학회지
• /
• 제35권1호
• /
• pp.1-18
• /
• 2022

This paper presents a statistical machine learning method that generates the implied volatility surface under the rareness of the market data. We apply the practitioner's Black-Scholes model and Gaussian process regression method to construct a Bayesian inference system with observed volatilities as a prior information and estimate the posterior distribution of the unobserved volatilities. The variance instead of the volatility is the target of the estimation, and the radial basis function is applied to the mean and kernel function of the Gaussian process regression. We present two types of Gaussian process regression methods and empirically analyze them.

• Computers and Concrete
• /
• 제24권1호
• /
• pp.7-17
• /
• 2019

The complex phenomenon of the bond formation in geopolymer is not well understood and therefore, difficult to model. This paper present applied statistical models for evaluating the compressive strength of geopolymer. The applied statistical models studied are divided into three different categories - linear regression [least absolute shrinkage and selection operator (LASSO) and elastic net], tree regression [decision and bagging tree] and kernel methods (support vector regression (SVR), kernel ridge regression (KRR), Gaussian process regression (GPR), relevance vector machine (RVM)]. The performance of the methods is compared in terms of error indices, computational effort, convergence and residuals. Based on the present study, kernel based methods (GPR and KRR) are recommended for evaluating compressive strength of Geopolymer concrete.

• 대한수학회논문집
• /
• 제12권3호
• /
• pp.797-812
• /
• 1997

For analytic functions we give an expression for the kernel $K_n$ of the remainder terms for the Gauss-Radau and the Gauss-Lobatto rules with end points of multiplicity r and prove the convergence of the kernel we obtained. The error bound are obtained for the type $$\mid$R_n(f)$\mid$ \leq \frac{1}{\pi}l(\Gamma) max_{z \in \Gamma} $\mid$K_n(z)$\mid$ max_{z \in \Gamma} $\mid$f(z)$\mid$$, where $l(\Gamma)$ denotes the length of contour $\Gamma$.