Combination of Multi-class Probability Support Vector Machines for Fault Diagnosis

To deal with multi-source multi-class classification problems, the method of combining multiple multi-class probability support vector machines （MPSVMs） using Bayesian theory is proposed in this paper. The MPSVMs are designed by mapping the output of standard support vector machines into a calibrated posterior probability by using a learned sigmoid function and then combining these learned binary-class probability SVMs. Two Bayes based methods for combining multiple MPSVMs are applied to improve the performance of classification. Our proposed methods are applied to fault diagnosis of a diesel engine. The experimental results show that the new methods can improve the accuracy and robustness of fault diagnosis.     

Support Vector Machine Ensemble Based on Genetic Algorithm

Support vector machines （SVMs） have been introduced as effective methods for solving classification problems. However, due to some limitations in practical applications, their generalization performance is sometimes far from the expected level. Therefore, it is meaningful to study SVM ensemble learning. In this paper, a novel genetic algorithm based ensemble learning method, namely Direct Genetic Ensemble （DGE）, is proposed. DGE adopts the predictive accuracy of ensemble as the fitness function and searches a good ensemble from the ensemble space. In essence, DGE is also a selective ensemble learning method because the base classifiers of the ensemble are selected according to the solution of genetic algorithm. In comparison with other ensemble learning methods, DGE works on a higher level and is more direct. Different strategies of constructing diverse base classifiers can be utilized in DGE. Experimental results show that SVM ensembles constructed by DGE can achieve better performance than single SVMs, hagged and boosted SVM ensembles. In addition, some valuable conclusions are obtained.     

Predicting and Classifying User Identification Code System Based on Support Vector Machines

In digital fingerprinting, preventing piracy of images by colluders is an important and tedious issue. Each image will be embedded with a unique User IDentification （UID） code that is the fingerprint for tracking the authorized user. The proposed hiding scheme makes use of a random number generator to scramble two copies of a UID, which will then be hidden in the randomly selected medium frequency coefficients of the host image. The linear support vector machine （SVM） will be used to train classifications by calculating the normalized correlation （NC） for the 2class UID codes. The trained classifications will be the models used for identifying unreadable UID codes. Experimental results showed that the success of predicting the unreadable UID codes can be increased by applying SVM. The proposed scheme can be used to provide protections to intellectual property rights of digital images aad to keep track of users to prevent collaborative piracies.     

Generating Fuzzy Rule-based Systems from Examples Based on Robust Support Vector Machine

This paper firstly proposes a new support vector machine regression （SVR） with a robust loss function, and designs a gradient based algorithm for implementation of the SVR,then uses the SVR to extract fuzzy rules and designs fuzzy rule-based system. Simulations show that fuzzy rule-based system technique based on robust SVR achieves superior performance to the conventional fuzzy inference method, the proposed method provides satisfactory performance with excellent approximation and generalization property than the existing algorithm.     

Support vector machine applied in QSAR modelling

Support vector machine （SVM）, partial least squares （PLS）, and Back-Propagation artificial neural network （ANN） were employed to establish QSAR models of 2 dipeptide datasets. In order to validate predictive capabilities on external dataset of the resulting models, both internal and external validations were performed. The division of dataset into both training and test sets was carried out by D-optimal design. The results showed that support vector machine （SVM） behaved well in both calibration and prediction. For the dataset of 48 bitter tasting dipeptides （BTD）, the results obtained by support vector regression （SVR） were superior to that by PLS in both calibration and prediction. When compared with BP artificial neural network, SVR showed less calibration power but more predictive capability. For the dataset of angiotensin-converting enzyme （ACE） inhibitors, the results obtained by support vector machine （SVM） regression were equivalent to those by PLS and BP artificial neural network. In both datasets, SVR using linear kernel function behaved well as that using radial basis kernel function. The results showed that there is wide prospect for the application of support vector machine （SVM） into QSAR modeling.     

PSVM,LSVM和NLSVM三种SVM分类算法的比较

支持向量机(SVM)是数据分类的强大工具,本文对三个分类算法进行了比较.这三个算法是最近SVM(PSVM),Lagrangian SVM(LSVM)和有限牛顿LSVM(NLSVM),比较了三个算法给出线性分类器的过程以及算法的速度和精度,提供了用SVM方法分类问题时的导向.     

基于双支持向量机的偏二叉树多类分类算法

提出一种基于双支持向量机的偏二叉树多类分类算法,偏二叉树双支持向量机多类分类算法.该算法综合了二叉树支持向量机和双支持向量机的优势,实现了在不降低分类性能的前提下,大大缩短训练时间.理论分析和UCI（University of California Irvine）机器学习数据库数据集上的实验结果共同证明,偏二叉树双支持...     

支持向量机方法中加权后验概率建模方法

为解决传统支持向量机方法不提供概率输出的问题,在支持向量机多类分类问题输出概率建模中,提出了加权后验概率建模方法。该方法在对多个两类支持向量机分类器的输出概率进行组合时,充分考虑了各个两类支持向量机分类器的差异。依据Bayesian理论,采用后验概率作为各个两类支持向量机分类器的权系数。实验结果表明,与投票法及Pairwise Coupling方法相比,加权后验概率方法具有较低的分类错误率,不仅具有较好的分类性能,而且得到的后验概率具有较好的概率分布形态。该方法有效地解决了实际多类分类问题中支持向量机的概率建模问题。     

基于层次型支持向量机的人脸检测

复杂背景中的人脸检测可广泛应用于人脸识别、人机交互等方面。但目前大部分人脸检测方法中存在分类器训练困难和检测计算量大等问题。提出了一种基于层次型支持向量机的正面直立人脸检测方法,在这两方面作了改进。这种结构的分类器由一个线性支持向量机组合和一个非线性支持向量机组成,由前者在保证检测率的情况下快速排除掉图像中绝大部分非人脸区域,后者对人脸候选区域进行进一步确认。在卡内基梅隆CMU等数据库上的实验证明了这种方法不仅具有较高的检测率和较低的误检率,而且具有较小的计算量。     

支持向量机的训练算法

大量数据下支持向量机(SVM)的训练算法是SVM研究的一个重要方向和广大研究者关注的焦点。该文回顾了近几年来这一领域的研究情况。该文从分析SVM训练问题的实质和难点出发,结合目前一些主要的SVM训练方法及它们之间的联系,重点阐述当前最有代表性的一种算法——序贯最小优化(SMO)算法及其改进算法。从中可以看到,包括SMO在内的分解算法通过求解一系列规模较小的子问题逐步逼近最优解,从而避免存储整个Hessian矩阵,是解决大规模SVM训练问题的主要方法。而工作集的选择对于分解算法的收敛与否和收敛速度至关重要。     

混沌时间序列的最小二乘支持向量机预测

提出了最小二乘支持向量机混沌时间序列预测方法,并研究了三种混沌信号的预测性能。该方法在优化指标中采用了平方项,且只有等式约束,将传统支持向量机求解二次规划问题转化为求解线性方程组,因而简化了计算复杂性。仿真实验结果表明该方法预测模型参数选择容易、在较大范围内取值时对预测误差影响很小,而且即使在输入维数m小于Takens嵌入定理所确定的维数时,也具有很好的预测性能。     

一种改进的快速支持向量机分类算法研究

快速的支持向量机分类算法--FCSVM对支持向量集采用变换的方式,用支持向量集的子集代替全部支持向量进行分类计算,在保证不损失分类精度的前提下使得分类速度较传统SVM算法有较大提高. 为了获得最小的支持向量子集,同时避免支持向量的移动,对FCSVM算法进行了改进.采用二分法优化分类函数中的支持向量数,给出了变换矩阵存在的充要条件及构造方法,减少了计算量.实验结果表明,改进的快速分类算法较大幅度地减少了计算复杂度,提高了分类速度,尤其在训练集规模庞大、支持向量数量较多的情况下,效果更加明显.     

支持向量机方法在线条痕迹检验中的应用

应用支持向量机的基本原理和信息融合思想,经对两种不同类型枪支所发射同一型号子弹上磨损线条状痕迹数据提取特征后,用特征小波包维数和磨损痕迹中坡膛痕迹尺度参数作为样本数据的因子分量,建立了较为理想的SVM分类模型,以此SVM分类模型实现了对磨损线条状痕迹的分类识别。研究表明：采用参数融合要比使用单一参数更有利于分类识别,并能提高分类识别率。     

一种支持向量集合的精简算法

由支持向量机算法得到的支持向量集合通常不是分类所必需的最小集合,冗余支持向量的存在降低了支持向量机的分类速度和实用化能力．为此,提出一种精简支持向量集合的新方法,给出了从原支持向量集合中识别和剔除冗余向量、生成新支持向量集合并确定其元素权值的算法．新方法尤其适用于样本规模大、支持向量数目多的分类问题．实验表明它能够在基本不降低支持向量机分类精度的前提下,大幅度地减少支持向量的数目,提高支持向量机的分类速度．     

降维近似支持向量机基因芯片数据分类器

基因芯片技术的出现改变了生物医学研究的前景,其产生的海量数据是限制其发展的瓶颈问题。论文针对基因芯片数据量大、样本数低和基因维数高的特点,提出了一种对基因芯片数据进行分类的降维近似支持向量机DRPSVM基因芯片数据分类器。DRPSVM采用降维的二次规划算法,使得该算法的时间复杂度和空间复杂度比传统的PSVM算法均有降低。通过在CAMDA2000、colon 1 dataset和colon 2 dataset等基因芯片数据集上的与BP、Nearest、RBF、SVM分类器的分类性能比较,DRPSVM在数据样本少、数据维数急剧升高时,分类性能稳定、存在唯一的最优解、训练时间快,适合基因芯片数据分类的应用环境。     

用于不平衡数据分类的模糊支持向量机算法

作为一种有效的机器学习技术,支持向量机已经被成功地应用于各个领域.然而当数据不平衡时,支持向量机会产生次优的分类模型;另一方面,支持向量机算法对数据集中的噪声点和野点非常敏感.为了克服以上不足,提出了一种新的用于不平衡数据分类的模糊支持向量机算法.该算法在设计样本的模糊隶属度函数时,不仅考虑训练样本到其类中心距离,而且考虑样本周围的紧密度.实验结果表明,所提模糊支持向量机算法可以有效地处理不平衡和噪声问题.