Location:
Search - libsvm weight
Search list
Description: 一种基于局部密度比权重设置模型的加权支持向量回归模型来单步求解多分类问题：该方法先分别对类样本中每类样本利用局部密度比权重设置模型求出每个样本的权重隶属因子,然后运用加权lib支持向量回归算法对所有样本进行训练,获得回归分类器,希望对大家有用!
Platform: |
Size: 41866 |
Author: 郭大 |
Hits:
Description: 一种基于局部密度比权重设置模型的加权支持向量回归模型来单步求解多分类问题：该方法先分别对类样本中每类样本利用局部密度比权重设置模型求出每个样本的权重隶属因子,然后运用加权lib支持向量回归算法对所有样本进行训练,获得回归分类器,希望对大家有用!-Based on local density than the right to re-set model of weighted support vector regression models to single-step to solve multi-classification problems: The method is first on the type of samples, respectively, for each type of samples the use of local density than the right to re-set the model to derive the right of each sample weight attached to factor, and then use lib weighted support vector regression algorithm for all samples for training, access to regression classifier, hope useful for all of us!
Platform: |
Size: 41984 |
Author: 郭大 |
Hits:
Description: 数据属性的权重分析.
用户可以给每个数据实例权重-Weights for data instances
Users can give a weight to each data instance.
Platform: |
Size: 49152 |
Author: quarryhero |
Hits:
Description: 一个实例搞定libsvm分类，实例问题如下以：一个班级里面有两个男生（男生1、男生2），两个女生（女生1、女生2）
• 男生1 身高：176cm 体重：70kg；
男生2 身高：180cm 体重：80kg；
女生1 身高：161cm 体重：45kg；
女生2 身高：163cm 体重：47kg；-Libsvm get an instance of the classification, examples of problems are to: a class which has two boys (boys 1 Boys 2), two girls (girls 1 girls 2) • Boys 1 Height: 176cm Weight: 70kg Boys 2 Height: 180cm Weight: 80kg girls 1 Height: 161cm Weight: 45kg girls 2 Height: 163cm Weight: 47kg
Platform: |
Size: 499712 |
Author: Jamei |
Hits:
Description: weight coding for support vector machine
Platform: |
Size: 107520 |
Author: Sakinah
|
Hits:
Description: 态势要素获取作为整个网络安全态势感知的基础，其质量的好坏将直接影响态势感知系统的性能。针对态势要素不易获取问题，提出了一种基于增强型概率神经网络的层次化框架态势要素获取方法。在该层次化获取框架中，利用主成分分析(PCA)对训练样本属性进行约简并对特殊属性编码融合处理，将其结果用于优化概率神经网络(PNN)结构，降低系统复杂度。以PNN作为基分类器，基分类器通过反复迭代、权重更替，然后加权融合处理形成最终的强多分类器。实验结果表明，该方案是有效的态势要素获取方法并且精确度达到95.53%，明显优于文中其他算法，有较好的泛化能力。(As the basis of the whole network security situation awareness, the quality of situation elements extraction will directly affect the performance of the situation awareness system. To solve the problem that the situation element is difficult to extract, we propose a method to extract the hierarchical frame situation elements based on the enhanced probabilistic neural network. In the hierarchical access frame, we use the principal component analysis (PCA) to reduct the training sample attribute and to process the special attribute encoding fusion. The result can be used to optimize the structure of the probabilistic neural network (PNN) and reduce the system complexity. Take PNN as the base classifier to form the final strong classifier by repeated iteration, weight replacement and weighted fusion. The experimental results show that the scheme is an effective method to obtain the situation factors and its accuracy is 95.53%,which is significantly better than other algorithms.)
Platform: |
Size: 1213440 |
Author: 莫言婷婷
|
Hits:
Description: 为了真实有效地提取网络安全态势要素信息，提出了一种基于增强型概率神经网络的层次化框架态势要素获取方法。在该层次化态势要素获取框架中，根据Agent节点功能的不同，划分为不同的层次。利用主成分分析(Principal Component Analysis, PCA)对训练样本属性进行约简并对特殊属性编码融合处理，按照处理结果改进概率神经网络(Probabilistic Neural Network, PNN)结构，以降低系统复杂度。然后以改进的PNN作为基分类器，结合自适应增强算法，通过基分类器反复迭代、样本权重更新，最后加权融合处理形成最终的强多分类器。实验结果表明，本文模型较文中其他几种方法具有较高的获取准确率和良好的泛化能力。(Firstly, in order to extract the information of network security situation accurately and effectively, a hierarchical frame feature acquisition method based on enhanced probabilistic neural network is proposed. According to different functions of Agent node, the hierarchical feature acquisition framework is divided into different levels. The principal component analysis (PCA) is used to reduce the training sample attributes and the special attribute encoding fusion. The result can be used to optimize the structure of the probabilistic neural network (PNN) so as to reduce the system complexity. Then, the improved PNN is used as the base classifier. Combined with the adaptive enhancement algorithm, the final strong classifier is formed through repeated iteration, weight replacement and weighted fusion. The experimental results show that the proposed model achieve higher accuracy and better generalization ability than other methods.)
Platform: |
Size: 98304 |
Author: 莫言婷婷
|
Hits: