2008-07-02
核选择是支持向量机研究中的核心问题之一,不同的核函数将产生不同的分类效果。本文研究了核参数和误差惩罚参数对支持向量机推广能力的影响,然后根据局部核函数与全局核函数的各自优点,提出了一种新的自适应组合
2011-03-31
Bergman核函数一直以来都是一个被大量讨论与研究的热门话题。由Bergman核函数所展开的一系列问题的研究也取得了很多成果。以前已经研究得到了Green函数与Bergman核函数之间的关系,并且
In this paper a method of the kernel select for the relevance vector machine is proposed. Former researches of RVM usually focus on the advantages of the algorithm in sparsely for less opportunities of over learning, how to accelerate the algorithm in order for needs such as large scale sets, and how to improve the precision of the algorithm by the combination of other algorithms such as PCA or SVM. The parameters of the kernel functions are regarded more important than these functions themselves, so that how to select the appropriate kernel and its parameters is confused for the application of RVM. In this paper, a method for kernel select for the RVM is proposed. Firstly, by simulation experiment, the relationship between σ in Gaussian kernel function and the regression result is discovered. Secondly, by analysis the reason of the experimental result, a common principle for kernel select is presented. Finally, on the principle, a composite kernel and the relation between different parameters and the number of RVs and the error rate in regression are tested by simulation experiment.
2013-12-19
School of Automation, Huazhong University of Sci. & Tech., Wuhan ,School of Automation, Huazhong University of Sci. & Tech., Wuhan
2005-02-07
应用并不多,而且核函数的选择也普遍采用径向基核,对于特殊核的应用研究很少。本文针对信号处理中的回归问题,研究了基于傅立叶核函数支持向量机的性能,并分析了参数q对其性能的影响;根据推导得出傅立叶核函数在
2015-04-21
人眼疲劳检测是预防驾驶员疲劳驾驶的有效途径。为了提高人眼疲劳检测的正确率,提出了一种基于混合核函数支持向量机的人眼疲劳检测方法。Adaboost算法进行人脸检测,引入灰度投影法准确定位人眼。使用
南京工业大学电子与信息工程学院,南京 211816,南京工业大学电子与信息工程学院,南京 211816
#计算机科学技术#
Support Vector Machine is a kind of algorithm used for classifying linear and nonlinear data, which not only has a solid theoretical foundation, but is more accurate than other sorting algorithms in many areas of applications, especially in dealing with high-dimensional data. It is not necessary for us to get the specific mapping function in solving quadratic optimization problem of SVM, and the only thing we need to do is to use kernel function to replace the complicated calculation of the dot product of the data set, reducing the number of dimension calculation. This paper introduces the theoretical basis of support vector machine, summarizes the research status and analyses the research direction and development prospects of kernel function.
2013-07-08
National Natural Science Foundation of China under Grant (No. 61172072)
the Beijing Natural Science Foundation under Grant ( 61271308)
the Beijing Science and Technology Program under Grant (No. 4112045)
the Research Fund for the Doctoral Program of Higher Education of China under Grant (No. Z121100000312024)
School of Electronic and Information Engineering, Beijing Jiaotong University, Beijing 100044,School of Electronic and Information Engineering, Beijing Jiaotong University, 100044,China Information Technology Security Evaluation Center, 100085
2017-01-31
时频分析方法由于能够兼顾信号的时频二维特征而在近年得到广泛使用,但在对多分量信号进行时频分析时会产生交叉项干扰。利用熵测度方法优化时频分布核函数的各项参数可以有效减少交叉项干扰,提高时频分布的精确度
北京邮电大学信息与通信工程学院,100876,北京邮电大学信息与通信工程学院,100876
2012-01-18
蛋白质关系实例结构化表达形式。在多个语料库上的实验表明,本文提出的PPI抽取方法其性能显著高于其它几种结构化信息表达形式。特别是,在AIMed语料上的PPI抽取取得了58.1的F1值,是目前基于单一核函数
江苏省自然科学基金(BK2010219)
教育部博士点基金(200802850006)
苏州大学计算机科学与技术学院,江苏苏州 215006,苏州大学计算机科学与技术学院,江苏苏州 215006,苏州大学计算机科学与技术学院,江苏苏州 215006,苏州大学计算机科学与技术学院,江苏苏州 215006
#计算机科学技术#
2012-01-04
本文提出了一种基于改进核函数的Wigner分布谱分解方法。即首先采用模糊域核函数滤波法抑制Wigner分布的交叉项,并在分析了指数型核函数后,给出了改进的新核函数。通过对新的核函数交叉项抑制能力的
高等学校博士学科点专项科研基金(200801830042)
吉林大学地球探测科学与技术学院,长春 130026,吉林大学地球探测科学与技术学院,长春 130026;大庆油田有限责任公司,大庆 163453,大庆油田有限责任公司,大庆 163453,吉林大学地球探测科学与技术学院,长春 130026
#地球科学#
2007-03-15
高分辨率面波成像应克服射线理论的高频近似假设,考虑有限频率效应。本文给出了一种计算面波有限频率3-D灵敏度核函数的方法。该方法可直接计算球壳层状地球模型中面波振型特征值和特征函数,而不再需要采用球化