不变矩的改进支持向量机在显微目标识别中的应用研究

为完成多个微小零件的识别,提出了一种改进的支持向量机分类算法．该算法应用基于边缘提取的不变矩获得特征属性,利用基于粗糙集的可辨识矩阵的启发式属性约简算法获得特征属性的约简,最后应用支持向量机进行目标识别分类．比较了使用支持向量机分类和使用提出的改进支持向量机分类对多个微小零件识别的效果．在显微视觉环境下的实验表明,提出的改进支持向量机分类方法能满足系统应用要求,分辨率达95%．     

基于增量学习支持向量机的音频例子识别与检索

音频例子识别与检索的主要任务是构造一个良好的分类学习机，而在构造过程中，从含有冗余样本的训练库中选择最佳训练例子、节省学习机的训练时间是构造分类机面临的一个挑战，尤其是对含有大样本训练库音频例子的识别．由于支持向量是支持向量机中的关键例子，提出了增量学习支持向量机训练算法．在这个算法中，训练样本被分成训练子库按批次进行训练，每次训练中，只保留支持向量，去除非支持向量．与普通和减量支持向量机对比的实验表明，算法在显著减少训练时间前提下，取得了良好的识别检索正确率．     

基于分层并行筛选样本的SVM增量学习算法

在增量学习过程中，随着训练集规模的增大，支持向量机的学习过程需要占用大量内存，寻优速度非常缓慢。在现有的一种支持向量机增量学习算法的基础上，结合并行学习思想，提出了一种分层并行筛选训练样本的支持向量机增量学习算法。理论分析和实验结果表明：与原有的算法相比，新算法能在保证支持向量机的分类能力的前提下显著提高训练速度。     

增量式约简最小二乘孪生支持向量回归机

为了解决增量式最小二乘孪生支持向量回归机存在构成的核矩阵无法很好地逼近原核矩阵的问题,提出了一种增量式约简最小二乘孪生支持向量回归机(IRLSTSVR)算法。该算法首先利用约简方法,判定核矩阵列向量之间的相关性,筛选出用于构成核矩阵列向量的样本作为支持向量以降低核矩阵中列向量的相关性,使得构成的核矩阵能够更好地逼近原核矩阵,保证解的稀疏性。然后通过分块矩阵求逆引理高效增量更新逆矩阵,进一步缩短了算法的训练时间。最后在基准测试数据集上验证算法的可行性和有效性。实验结果表明,与现有的代表性算法相比,IRLSTSVR算法能够获得稀疏解和更接近离线算法的泛化性能。     

基于核矩阵学习的XML文档相似度量方法

XML文档作为一种新的数据形式,成为当前的研究热点.XML文档间相似度的计算是XML文档分析、管理及文本挖掘的基础.结构链接向量模型(structuredlink vector model,简称SLVM)是一种综合考虑XML文档结构信息与内容信息进行XML文档相似度量的方法.体现XML文档结构单元关系的核矩阵在结构链接向量模型中扮演着重要角色.为自动捕获XML文档结构单元关系,提出了两种核矩阵的学习算法,分别是基于支持向量机(support vector machine,简称SVM)的回归学习算法和基于矩阵迭代的学习算法.相似搜索实验对比结果表明,基于核矩阵学习方法的XML文档相似度量方法的准确性明显优于其他方法.进一步实验表明,基于矩阵迭代学习的核矩阵学习算法与基于支持向量机的回归学习算法相比,不仅具有更高的准确性,而且所需训练文档更少、计算代价更小.     

基于支持向量机的货币识别应用研究

支持向量机是一种新的基于统计学习理论的机器学习算法，它可以应用于小样本、非线性和高维模式识别。研究了支持向量机的学习算法，依据支持向量机的特点采用了相应的货币特征数据获取及预处理方法，提出采用改进SMO训练算法和DAGSVM多值分类算法构建的支持向量机用于货币识别，从而达到对货币高效、准确识别。实验结果证实了该方案的有效性。     

一种适合于增量学习的支持向量机的快速循环算法

当样本数量大到计算机内存中放不下时，常规支持向量机方法就失去了学习能力，为了解决这一问题，提高支持向量机的训练速度，文章分析了支持向量机分类的本质特征，根据支持向量机分类仅与支持向量有关的特点，提出了一种适合于支持向量机增量学习的快速循环算法(PFI-SVM)，提高了支持向量机的训练速度和大样本学习的能力，而支持向量机的分类能力不受任何影响，取得了较好的效果。     

基于统计学习理论的支持向量机的分类方法

支持向量机是一种新型机器学习方法，由于其出色的学习性能，该技术已成为机器学习领域新的研究热点。介绍用于分类的支持向量机的统计学习理论基础，在此基础上提出了支持向量机的分类算法，讨论了支持向量机存在的问题，对用于分类的支持向量机的应用前景进行了展望。     

支持向量机的新发展

Vapnik等学者首先提出了实现统计学习理论中结构风险最小化原则的实用算法一支持向量机，比较成功地解决了模式分类问题，其后，机器学习界兴起了研究统计学习理论和支持向量机的热湖，引人瞩目的研究分支有从最优化技术出发改进或改造支持向量机，依据统计学习理论和支持向量机的优点设计新的非线性机器学习算法等，对此，较为系统地回顾了近lO年来算法研究领域的新发展。     

支持向量机组合分类及其在文本分类中的应用

针对标准支持向量机对野值点和噪音敏感，分类时明显倾向于大类别的问题，提出了一种同时考虑样本差异和类别差异的双重加权支持向量机。并给出了由近似支持向量机结合支持向量识别算法，识别野值点和计算样本重要性权值的方法．双重加权支持向量机和近似支持向量机组合的新分类算法尤其适用于样本规模大、样本质量不一、类别不平衡的文本分类问题．实验表明新算法改善了分类器的泛化性能。比传统方法具有更高的查准率和查全率．     

USING LEAST SQUARE SVM FOR NONLINEAR SYSTEMS MODELING AND CONTROL

Support vector machine is a learning technique based on the structural risk minimization principle, and it is also a class of regression method with good generalization ability. The paper firstly introduces the mathematical model of regression least squares support vector machine (LSSVM), and designs incremental learning algorithms by the calculation formula of block matrix, then uses LSSVM to model nonlinear system, based on which to control nonlinear systems by model predictive method. Simulation experiments indicate that the proposed method provides satisfactory performance, and it achieves superior modeling performance to the conventional method based on neural networks, moreover it achieves well control performance.     

Statistical learning algorithms for identifying contrasting tillage practices with Landsat Thematic Mapper data

Tillage management practices have a direct impact on water-holding capacity, evaporation, carbon sequestration and water quality. This study examines the feasibility of two statistical learning algorithms, namely the least square support vector machine (LSSVM) and relevance vector machine (RVM), for identifying two contrasting tillage management practices using remote-sensing data. LSSVM is firmly based on statistical learning theory, whereas RVM is a probabilistic model where the training takes place in a Bayesian framework. Input to the LSSVM and RVM algorithms were reflectance values at different bandwidths and indices derived from Landsat Thematic Mapper (TM) data. Ground-truth data for this study were collected from 72 commercial production fields in two counties located in the Texas High Plains of the south-central USA. Numerous LSSVM- and RVM-based tillage models were developed and evaluated for tillage classification accuracy. The percentage correct and kappa statistic were used for the evaluation. The results showed that the best LSSVM and RVM models included the use of TM band 5 or vegetation indices that involved TM band 5, indicating sensitivity of near-infrared reflectance of crop residue cover on the surface. This is consistent with other remote-sensing models reported in the literature. Overall classification accuracies of the best LSSVM and RVM models were 87.8 and 90.2%, respectively. The corresponding kappa statistics for those models were 0.75 and 0.80, respectively. Furthermore, comparison of the best LSSVM and RVM models with the published logistic regression-based tillage models developed with the same data indicated the superiority of the RVM model over LSSVM and logistic regression models in determining contrasting tillage practices with Landsat TM data.     

一种实用的火电厂飞灰含碳量软测量建模方法

提出了同时利用自适应加权融合和最小二乘支持向量机建模的实用新方法。首先,给出了基于小波的自适应加权融合和最小二乘支持向量机算法;其次,将BP神经网络、最小二乘支持向量机和基于小波的自适应加权融合的最小二乘支持向量机算法进行建模精度比较;最后,采用真实火电厂飞灰含碳量数据进行模型验证与预测,仿真结果表明基于小波的自适应加权融合的最小二乘支持向量机算法具有较好的建模精度和实用性。     

基于信息融合的火电厂飞灰含碳量的软测量建模实验室

提出了利用信息融合与软测量技术对火电厂飞灰含碳量测量进行建模的新算法．首先给出了自适应加 权融合和最小二乘支持向量机（LSSVM）算法,其次对三个非线性测试函数分别运用BP 神经网络、LSSVM 和基于 自适应加权融合的LSSVM 算法进行建模并比较了精度,最后给出了基于自适应加权融合的LSSVM 在火电厂飞灰 含碳量建模中应用的结果．     

基于- 范数约束的LSSVR 多核学习算法

针对核函数选择对最小二乘支持向量机回归模型泛化性的影响, 提出一种新的基于????- 范数约束的最小二乘支持向量机多核学习算法. 该算法提供了两种求解方法, 均通过两重循环进行求解, 外循环用于更新核函数的权值, 内循环用于求解最小二乘支持向量机的拉格朗日乘数, 充分利用该多核学习算法, 有效提高了最小二乘支持向量机的泛化能力, 而且对惩罚参数的选择具有较强的鲁棒性. 基于单变量和多变量函数的仿真实验表明了所提出算法的有效性.

     

Least square support vector machine and multivariate adaptive regression spline for modeling lateral load capacity of piles

This article adopts least square support vector machine (LSSVM) and multivariate adaptive regression spline (MARS) for prediction of lateral load capacity (Q) of pile foundation. LSSVM is firmly based on the theory of statistical learning, uses regression technique. MARS is a nonparametric regression technique that models complex relationships. Diameter of pile (D), depth of pile embedment (L), eccentricity of load (e), and undrained shear strength of soil (S _u) have been used as input parameters of LSSVM and MARS. Equations have been presented from the developed MARS and LSSVM. This study also presents a comparative study between the developed MARS and LSSVM.     

一种改进的SVM回归估计算法

支持向量机是一种基于结构风险最小化原理的学习技术,也是一种新的具有很好泛化性能的回归方法。目前,如何设计快速有效的回归估计算法仍然是支持向量机实际应用中的问题之一。文中对标准SVM回归估计算法加以改进,提出一种改进的SVM回归估计算法,并从学习速度和回归估计精度两个方面对提出的改进的SVM回归估计算法与标准SVM回归估计算法进行了比较。实验结果表明,在学习速度与回归估计精度之间取折衷时,文中提出的回归估计算法自由度更大。     

A Hybrid Heat Rate Forecasting Model Using Optimized LSSVM Based on Improved GSA

Heat rate value is considered as one of the most important thermal economic indicators, which determines the economic, efficient and safe operation of steam turbine unit. At the same time, an accurate heat rate forecasting is core task in the optimal operation of steam turbine unit. Recently, least squares support vector machine (LSSVM) is being proved an effective machine learning technique for solving nonlinear regression problem with a small sample set. However, it has also been proved that the prediction precision of LSSVM is highly dependent on its parameters, which are hardly choosing for the LSSVM. In the paper, an improved gravitational search algorithm (AC-GSA) is presented to further enhance optimal performance of GSA, and it is employed to serve as an approach for pre-selecting LSSVM parameters. Then, a novel soft computing method, based on LSSVM and AC-GSA, is therefore proposed to forecast heat rate of a 600 MW supercritical steam turbine unit. It combines the merits of the high accuracy of LSSVM and the fast convergence of GSA in order to build heat rate prediction model and obtain a well-generalized model. Results indicate that the developed AC-GSA–LSSVM model demonstrates better regression precision and generalization capability.     

An unbiased LSSVM model for classification and regression

Aiming at the common support vector machine’s biased disadvantage and computational complexity, an unbiased least squares support vector machine (LSSVM) model is proposed in this paper. The model eliminates the bias item of LSSVM by improving the form of structure risk, then the unbiased least squares support vector classifier and the unbiased least squares support vector regression are deduced. Based on this model, we design a new learning algorithm using Cholesky factorization according to the characteristic of kernel function matrix, in this way the calculation of Lagrangian multipliers is greatly simplified. Several experiments on diffenert datasets are carried out, including the common datasets classification, synthetic aperture radar image automatic target recognition and chaotic time series prediction. The experimental results of correct recognition rate and the fitting precision testify that the unbiased LSSVM model has good universal ability and fitting accuracy, better generalization capability and stability, and have a great improvement in learning speed.     

Optimal Locality Regularized Least Squares Support Vector Machine via Alternating Optimization

The least squares support vector machine (LSSVM), like standard support vector machine (SVM) which is based on structural risk minimization, can be obtained by solving a simpler optimization problem than that in SVM. However, local structure information of data samples, especially intrinsic manifold structure, is not taken full consideration in LSSVM. To address this problem and inspired by manifold learning technique, we propose a novel iterative least squares classifier, coined optimal locality preserving least squares support vector machine (OLP-LSSVM). The idea is to combine structural risk minimization and locality preserving criterion in a unified framework to take advantage of the manifold structure of data samples to enhance LSSVM. Furthermore, inspired by the recent development of simultaneous optimization technique, adjacent graph of locality preserving criterion is optimized simultaneously to give rise to improved discriminative performance. The resulting model can be solved by alternating optimization method. The experimental results on several publicly available benchmark data sets show the feasibility and effectiveness of the proposed method.