Optimization Algorithm with Kernel PCA to Support Vector Machines for Time Series Prediction

Chen, Qisong; Chen, Xiaowei; Wu, Yun

doi:10.4304/jcp.5.3.380-387

Cited by 15 publications

(6 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The calculated contact stresses of high-speed stage need to be less than the allowable stresses respectively, as shown in (3).…”

Section: Strength Constraint Conditionsmentioning

confidence: 99%

Optimization of Two-Stage Cylindrical Gear Reducer with Adaptive Boundary Constraints

Li¹,

Jiang²,

Zeng³

2013

JSW

View full text Add to dashboard Cite

An optimization mathematical model for designing two-stage cylindrical gear reducer is firstly constructed based upon the drive principles and design criteria of cylindrical gear transmission. Then some techniques for calculating the key performance parameters, such as profile factor and dynamic load coefficient, etc., and revising the corresponding boundary constraints dynamically during the optimization process are researched and implemented. Finally, dedicated interactive optimization design software is developed based on VC++/MATLAB mixed programming. Compared with the conventional optimization methods with fixed performance parameter, the proposed method can accurately calculate all the key parameters and revise the corresponding constraint functions adaptively in each iteration step. Optimization examples illustrate that the proposed gear optimization method is more effective and reasonable.

show abstract

“…The calculated contact stresses of high-speed stage need to be less than the allowable stresses respectively, as shown in (3).…”

Section: Strength Constraint Conditionsmentioning

confidence: 99%

Optimization of Two-Stage Cylindrical Gear Reducer with Adaptive Boundary Constraints

Li¹,

Jiang²,

Zeng³

2013

JSW

View full text Add to dashboard Cite

show abstract

“…Nonnegative Matrix Factorization (NMF) [16] is a recently developed technique for nonlinearly finding purely additive, parts-based, linear, and low-dimension representations of nonnegative multivariate data to consequently reveal the latent structure, feature or pattern in the data. Given a non-negative data matrix V, NMF finds an approximate factorization V into non-negative factors W and H. The non-negativity constraints make the representation purely additive (allowing no subtractions), in contrast to many other linear representations such as principal component analysis [17] (PCA) and independent component analysis (ICA). Also many extended NMF algorithms have been proposed.…”

Section: B Non-negative Matrix Factorizationmentioning

confidence: 99%

Software Defect Prediction Using Non-Negative Matrix Factorization

Chang¹,

Mu²,

Zhang³

2011

JSW

View full text Add to dashboard Cite

Quality is considered as an important issue in the fields of software engineering. However, building quality software is very expensive, in order to raise the effectiveness and efficiency of quality assurance and testing, software defect prediction is used to identify defect-prone modules in an upcoming version of a software system and help to allow the effort on those modules. Although many models have been proposed, this problem has not resolved thoroughly. For overcoming these limits, recent results show that researcher should pay more attention to improve the quality of the data. Aimed at this purpose, in this paper, we propose a novel approach to resolve the problem of software defect prediction. The method is classification using Non-Negative Matrix Factorization (NMF). In this paper, NMF algorithm is not only used for extracting external features but also as a powerful way for classification of software defect data. Experiments demonstrating the efficiency of the proposed approach are performed for software defect data classification. And the results show that it outperforms the state of the art techniques tested for this experiment. Finally, we suggest that it can be a useful and practical way addition to the framework of software quality prediction

show abstract

“…Firstly proposed by Vapnik et al, Support Vector Machine (SVM) is a machine learning method which is applied to solve the binary classification problem [1][2][3]. It is an algorithm based on the VC dimension theory and the principle of structural risk minimization in the statistical learning theory,and it has the features of optimization, nuclear and the best generalization ability [4,5].The majority of the scholars have been concerned about it and it has been applied in many fields in recent years [6][7][8][9][10][11]. The majority of researchers have proposed many improved algorithms on the basis of SVM.…”

Section: Introductionmentioning

confidence: 99%

Mixed Kernel Twin Support Vector Machines Based on the Shuffled Frog Leaping Algorithm

Wu¹,

Ding²,

Huang³

et al. 2014

JCP

View full text Add to dashboard Cite

The efficiency and performance of Twin Support Vector Machines (TWSVM) is better than the traditional support vector machines when it deals with the problems. However, it also has some problems. As the same as the traditional support vector machines, its parameters are difficult to be appointed and it is not easy to select the appropriate kernel function. TWSVM generally selects the Gaussian radial basis kernel function. Although its learning ability is very strong, its generalization ability is relatively weak. To a certain extent, this limits the performance of TWSVM .In order to solve these two problems, in this paper, we propose the Mixed Kernel Twin Support Vector Machines based on the shuffled frog leaping algorithm (SFLA-MK-TWSVM). To make full use of both the excellent generalization ability of global kernel functions and the learning ability of local kernel functions, SFLA-MK-TWSVM constructs a mixed kernel which has better performance. Then SFLA-MK-TWSVM uses the shuffled frog leaping algorithm to determine the parameters of both TWSVM and the mixed kernel function to further improve the performance of TWSVM. The experimental results indicate that SFLA-MK-TWSVM significantly improves the classification accuracy of TWSVM.Index Terms-the mixed kernel function, TWSVM, the shuffled frog leaping function

show abstract

Optimization Algorithm with Kernel PCA to Support Vector Machines for Time Series Prediction

Cited by 15 publications

References 21 publications

Optimization of Two-Stage Cylindrical Gear Reducer with Adaptive Boundary Constraints

Optimization of Two-Stage Cylindrical Gear Reducer with Adaptive Boundary Constraints

Software Defect Prediction Using Non-Negative Matrix Factorization

Mixed Kernel Twin Support Vector Machines Based on the Shuffled Frog Leaping Algorithm

Contact Info

Product

Resources

About