• Journal of applied mathematics & informatics
• /
• 제34권1_2호
• /
• pp.95-106
• /
• 2016

In this paper, according to the classical LSQR algorithm forsolving least squares (LS) problem, an iterative method is proposed for finding the minimum-norm pure imaginary solution of the quaternionic least squares (QLS) problem. By means of real representation of quaternion matrix, the QLS's correspongding vector algorithm is rewrited back to the matrix-form algorthm without Kronecker product and long vectors. Finally, numerical examples are reported that show the favorable numerical properties of the method.

• ETRI Journal
• /
• 제28권2호
• /
• pp.155-161
• /
• 2006

In this paper, we introduce an escalator (ESC) algorithm based on the least squares (LS) criterion. The proposed algorithm is relatively insensitive to the eigenvalue spread ratio (ESR) of an input signal and has a faster convergence speed than the conventional ESC algorithms. This algorithm exploits the fast adaptation ability of least squares methods and the orthogonalization property of the ESC structure. From the simulation results, the proposed algorithm shows superior convergence performance.

• Journal of the Korean Data and Information Science Society
• /
• 제27권2호
• /
• pp.549-557
• /
• 2016

In this paper we propose multiclassification method for large data sets by ensembling least squares support vector machines (LS-SVM) with principal components instead of raw input vector. We use the revised one-vs-all method for multiclassification, which is one of voting scheme based on combining several binary classifications. The revised one-vs-all method is performed by using the hat matrix of LS-SVM ensemble, which is obtained by ensembling LS-SVMs trained using each random sample from the whole large training data. The leave-one-out cross validation (CV) function is used for the optimal values of hyper-parameters which affect the performance of multiclass LS-SVM ensemble. We present the generalized cross validation function to reduce computational burden of leave-one-out CV functions. Experimental results from real data sets are then obtained to illustrate the performance of the proposed multiclass LS-SVM ensemble.

• Journal of the Korean Data and Information Science Society
• /
• 제23권4호
• /
• pp.739-748
• /
• 2012

최소제곱 서포트벡터기계 (least squares support vector machine)는 분류 및 비선형 회귀분석에서 유용하게 사용되고 있는 통계적 기법이다. 본 논문에서는 각 집단별로 생존자료가 관측된 경우 적용할 수 있는 LS-SVM을 제안한다. 제안된 모형은 임의우측 중도절단자료를 비선형 회귀모형에 적용할 수 있게 Kaplan- Meier의 중도절단분포의 추정값을 이용하여 구해진 가중값을 사용하고, 집단 간의 변동을 나타내기 위하여 임의효과항을 포함한다. 벌칙상수와 커널모수의 최적값을 구하기 위하여 일반화 교차타당성함수가 사용되고 모의실험에서는 임의효과항을 포함하지 않은 LS-SVM과 성능을 비교함으로써 제안된 방법의 우수성을 보이기로 한다.

• 한국음향학회지
• /
• 제15권4호
• /
• pp.93-97
• /
• 1996

완전최소자승법(total least squares method, TLS) Ax${\simeq}$b와 같은 형태의 시스템 식을 푸는데 있어 데이터 행렬 A와 b에 잡음비 섞인 경우에 편이 되지 않은 해를 구하기 위하여 널리 이용된다. 그러나 임펄수성의 잡음과 같은 heavy tailed 확률분포를 갖는 잡음이 존재할 때 완전 최소자승법은 unbiased estimator이지만 최소자승법(least squares, LS)과 마찬가지로 경인하지 못한 성능을 보인다. 본 논문에서는 TLS 방법의 견인성에 대하여 논하고 완전최소자승법의 해의 특성을 기반으로 하여 견인한 완전최소자승법(robust TLS, ROTLS)을 제안한다. 또한 ROTLS 방법을 시스템식별문제에 적용하여 그 성능을 평가한다.

• Journal of the Korean Data and Information Science Society
• /
• 제21권5호
• /
• pp.831-839
• /
• 2010

최소제곱 서포트벡터기계는 비선형회귀분석과 분류에 널리 쓰이는 커널기법이다. 본 논문에서는 금융시계열자료의 평균 및 변동성을 추정하기 위하여 평균의 추정 방법으로는 가중최소제곱 서포트벡터기계, 변동성의 추정 방법으로는 최소제곱 서포트벡터기계를 사용하는 비선형 평균 일반화 이분산 자기회귀모형을 제안한다. 제안된 모형은 선형 일반화 이분산 자기회귀모형 및 선형 평균 일반화 이분산 자기회귀모형보다 더 나은 추정 능력을 가진다는 것을 실제자료의 추정을 통하여 보였다.

• Structural Monitoring and Maintenance
• /
• 제5권3호
• /
• pp.379-390
• /
• 2018

In this paper, a novel and effective damage diagnosis algorithm is proposed to detect and estimate damage using two stages least squares support vector machine (LS-SVM) and limited number of attached sensors on structures. In the first stage, LS-SVM1 is used to predict the unmeasured mode shapes data based on limited measured modal data and in the second stage, LS-SVM2 is used to predicting the damage location and severity using the complete modal data from the first-stage LS-SVM1. The presented methods are applied to a three story irregular frame and cantilever plate. To investigate the noise effects and modeling errors, two uncertainty levels have been considered. Moreover, the performance of the proposed methods has been verified through using experimental modal data of a mass-stiffness system. The obtained damage identification results show the suitable performance of the proposed damage identification method for structures in spite of different uncertainty levels.

• Journal of the Korean Data and Information Science Society
• /
• 제14권2호
• /
• pp.337-343
• /
• 2003

In this paper we present the prediction interval estimation method using bootstrap method for least squares support vector machine(LS-SVM) regression, which allows us to perform even nonlinear regression by constructing a linear regression function in a high dimensional feature space. The bootstrap method is applied to generate the bootstrap sample for estimation of the covariance of the regression parameters consisting of the optimal bias and Lagrange multipliers. Experimental results are then presented which indicate the performance of this algorithm.

• 한국데이터정보과학회:학술대회논문집
• /
• 한국데이터정보과학회 2005년도 추계학술대회
• /
• pp.161-169
• /
• 2005

In this paper we propose a fuzzy c-regression model based on weighted least squares support vector machine(LS-SVM), which can be used to detect outliers in the switching regression model while preserving simultaneous yielding the estimates of outputs together with a fuzzy c-partitions of data. It can be applied to the nonlinear regression which does not have an explicit form of the regression function. We illustrate the new algorithm with examples which indicate how it can be used to detect outliers and fit the mixed data to the nonlinear regression models.

• Communications for Statistical Applications and Methods
• /
• 제18권6호
• /
• pp.817-835
• /
• 2011

The moment(MM) and least squares(LS) estimations of the parameters are derived for the Pareto distribution in the presence of outliers. Further, we have derived a mixture method(MIX) of estimations with MM and LS that shows that the MIX is more efficient. In the final section we have given an example of actual data from a medical insurance company.