• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2000년도 봄 학술발표논문집 Vol.27 No.1 (B)
• /
• pp.469-471
• /
• 2000

본 논문에서는 모멘트 항을 추가한 수정된 kernel-adatron 알고리즘을 제안하고 이른 support vector machines의 학습기법으로 이용하였다. 이는 기울기상승법에서 일어나는 최적해로의 수렴에 따른 발진을 억제하여 그 수렴 속도를 좀더 개선시키는 모멘트의 장점과 kernel-adatron 알고리즘의 구현용이성을 그대로 살리기 위함이다. 제안된 학습기법의 SVM을 실제 200명의 암환자를 2부류(초기와 악성)로 분류하여 문제에 적용하여 시뮬레이션한 결과, Cambell등의 kernel-adatron 알고리즘을 이용한 SVM의 결과와 비교할 때 학습시간과 시험 데이터의 분류률에서 더욱 우수한 성능이 있음을 확인할 수 있었다.

• Journal of Computing Science and Engineering
• /
• 제10권2호
• /
• pp.58-67
• /
• 2016

With the large amount of complex network data that is increasingly available on the Web, link prediction has become a popular data-mining research field. The focus of this paper is on a link-prediction task that can be formulated as a binary classification problem in complex networks. To solve this link-prediction problem, a sparse-classification algorithm called "Truncated Kernel Projection Machine" that is based on empirical-feature selection is proposed. The proposed algorithm is a novel way to achieve a realization of sparse empirical-feature-based learning that is different from those of the regularized kernel-projection machines. The algorithm is more appealing than those of the previous outstanding learning machines since it can be computed efficiently, and it is also implemented easily and stably during the link-prediction task. The algorithm is applied here for link-prediction tasks in different complex networks, and an investigation of several classification algorithms was performed for comparison. The experimental results show that the proposed algorithm outperformed the compared algorithms in several key indices with a smaller number of test errors and greater stability.

• Journal of the Korean Data and Information Science Society
• /
• 제19권4호
• /
• pp.1219-1231
• /
• 2008

The main object of this paper is to develop a leave-one-out(LOO) bound of all pairwise comparison error correcting output codes (APC-ECOC). To avoid using classifiers whose corresponding target values are 0 in APC-ECOC and requiring pilot estimates we developed a bound based on mean misclassification probability(MMP). It can be used to tune kernel hyperparameters. Our empirical experiment using kernel mean squared estimate(KMSE) as the binary classifier indicates that the bound leads to good estimates of kernel hyperparameters.

• 한국유통학회지:유통연구
• /
• 제10권3호
• /
• pp.1-14
• /
• 2005

성공적인 공급사슬관리에 있어 성과에 따른 지속적 협업 통제는 매우 중요하다. 본 연구에서는 기계학습 알고리즘인 SVM(Support Vector Machiness)을 이용해 균형성과표에 기반한 공급사슬관리 성과에 따른 지속적 협업 통제 모델을 개발하였다. 우리는 지속적 협업 통제모델 개발에 있어 108명의 전문가를 상대로 실증조사를 수행하였다. 본 연구 수행에 있어 4가지 형태의 SVM 커늘 (1) linear, (2) polynomail, (3) Radial Basis Function(RBF), (4) sigmoid kernel을 이용해 공급사슬관리 지속적 협업 예측 정확도를 비교하였다. SVM 커늘 4가지 중 linear kernel의 예측성과가 가장 좋았다. 그리고 본 연구에서는 SVM linear kernel의 예측성과를 ANN(Artificial Neural Network)의 예측성과와 비교하였다. 분석결과 SVM linear kernel이 공급사슬관리에 있어 지속적 협업 예측에 우수한 예측성과를 보여주는 것을 발견하였다. 이러한 곁과는 SVM linear kernel이 공급사슬관리의 지속적 협업 예측 통제에 있어 우수한 대안을 제공해 줄 것이다. 그러므로 공급사슬관리를 추구하는 기업들은 분 모델을 통해 지속적 협업 통제에 유용한 정보를 얻을 수 있을것이다.

• 한국경영정보학회:학술대회논문집
• /
• 한국경영정보학회 2007년도 International Conference
• /
• pp.608-613
• /
• 2007

Prediction of corporate bankruptcies has long been an important topic and has been studied extensively in the finance and management literature because it is an essential basis for the risk management of financial institutions. Recently, support vector machines (SVMs) are becoming popular as a tool for bankruptcy prediction because they use a risk function consisting of the empirical error and a regularized term which is derived from the structural risk minimization principle. In addition, they don't require huge training samples and have little possibility of overfitting. However. in order to Use SVM, a user should determine several factors such as the parameters ofa kernel function, appropriate feature subset, and proper instance subset by heuristics, which hinders accurate prediction results when using SVM In this study, we propose a novel hybrid SVM classifier with simultaneous optimization of feature subsets, instance subsets, and kernel parameters. This study introduces genetic algorithms (GAs) to optimize the feature selection, instance selection, and kernel parameters simultaneously. Our study applies the proposed model to the real-world case for bankruptcy prediction. Experimental results show that the prediction accuracy of conventional SVM may be improved significantly by using our model.

• 응용통계연구
• /
• 제26권1호
• /
• pp.131-139
• /
• 2013

지지벡터기계는 잡음변수가 존재하는 경우에 성능이 저하될 수 있다. 또한 최종 분류기에서 각 변수들의 중요도를 알리 어려운 단점이 있다. 따라서 변수선택은 지지벡터기계의 해석력과 정확도를 높일 수 있다. 기존의 문헌상의 대부분의 연구는 선형 지지벡터기계에서 성근 해를 주는 벌점함수를 통해 변수를 선택에 관한 것이다. 실제로는 분류의 정확도를 높이기 위해 비선형 커널을 사용하는 경우가 일반적이다. 따라서 변수선택은 비선형 지지벡터기계에서도 마찬가지로 필요하다. 본 논문에서는 모의실험 및 실제자료를 통하여 비선형 지지벡터의 대표적인 변수선택법인 COSSO(component selection and smoothing operator)와 KNIFE(kernel iterative feature extraction)의 성능을 비교한다.

• Journal of the Korean Data and Information Science Society
• /
• 제25권6호
• /
• pp.1539-1547
• /
• 2014

In this paper we apply the autoregressive process to the nonlinear quantile regression in order to infer nonlinear quantile regression models for the autocorrelated data. We propose a kernel method for the autoregressive data which estimates the nonlinear quantile regression function by kernel machines. Artificial and real examples are provided to indicate the usefulness of the proposed method for the estimation of quantile regression function in the presence of autocorrelation between data.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2003년도 봄 학술발표논문집 Vol.30 No.1 (B)
• /
• pp.175-177
• /
• 2003

This paper addresses a problem of classifying human breast cancer into its subtypes. A main ingredient in our approach is kernel machines such as support vector machine (SVM). kernel principal component analysis (KPCA). and kernel partial least squares (KPLS). In the task of breast cancer classification, we employ both SVM and KPLS and compare their results. In addition to this classification. we also analyze the patterns of clinical outcomes in the feature space. In order to visualize the clinical outcomes in low-dimensional space, both KPCA and KPLS are used. It turns out that these methods are useful to identify correlations between clinical outcomes and the nonlinearly protected expression profiles in low-dimensional feature space.

• Journal of Information Processing Systems
• /
• 제20권2호
• /
• pp.263-272
• /
• 2024

A model based on genetic algorithm optimization, GA-SVM, is proposed to warn university students of their status. This model improves the predictive effect of support vector machines. The genetic optimization algorithm is used to train the hyperparameters and adjust the kernel parameters, kernel penalty factor C, and gamma to optimize the support vector machine model, which can rapidly achieve convergence to obtain the optimal solution. The experimental model was trained on open-source datasets and validated through comparisons with random forest, backpropagation neural network, and GA-SVM models. The test results show that the genetic algorithm-optimized radial basis kernel support vector machine model GA-SVM can obtain higher accuracy rates when used for early warning in university learning.

• 한국통신학회논문지
• /
• 제34권12C호
• /
• pp.1177-1183
• /
• 2009

In this paper, support vector machines (SVM) are applied to multi-user detector (MUD) for direct sequence (DS)-CDMA system. This work shows an analytical performance of SVM based multi-user detector with some of kernel functions, such as linear, sigmoid, and Gaussian. The basic idea in SVM based training is to select the proper number of support vectors by maximizing the margin between two different classes. In simulation studies, the performance of SVM based MUD with different kernel functions is compared in terms of the number of selected support vectors, their corresponding decision boundary, and finally the bit error rate. It was found that controlling parameter, in SVM training have an effect, in some degree, to SVM based MUD with both sigmoid and Gaussian kernel. It is shown that SVM based MUD with Gaussian kernels outperforms those with other kernels.