Show simple item record

dc.contributor.advisor谢邦昌
dc.contributor.author程晓华
dc.date.accessioned2018-12-05T01:29:41Z
dc.date.available2018-12-05T01:29:41Z
dc.date.issued2017-11-02
dc.identifier.urihttps://dspace.xmu.edu.cn/handle/2288/169198
dc.description.abstract在运用支持向量机进行高维数据分类时,我们往往在支持向量机中加入某些惩罚以去除不相关的预测因子,提高预测准确率。Lasso和其他变量选择方法已成功应用到支持向量机中,达到自动进行变量选择的效果。但在许多实际问题中,简单的线性可加模型不能够很好的捕捉到预测因子与响应变量之间的关系,加入变量的交互项会使模型预测力增强。例如,在疾病诊断时,两种症状的同时发生会帮助医生做出更明确的判断;在寻找病因时,基因与基因、基因与环境因素的相互作用显得尤为重要。当存在交互项时,变量之间潜在存在着分层结构,Lasso等方法所得到的模型常常违背这种分层结构,使所得模型难以解释。因此本文在支持向量机中进行变量选择的同时施...
dc.description.abstractWhen applying support vector machine to high-dimension data classification, we often add some penalty to the support vector machine to remove the irrelevant predictors. Lasso and other variable selection methods have been successfully applied to the support vector machine, which can perform variable selection automatically. But in many practical problems, the simple linear additive model can not c...
dc.language.isozh_CN
dc.relation.urihttps://catalog.xmu.edu.cn/opac/openlink.php?strText=57040&doctype=ALL&strSearchType=callno
dc.source.urihttps://etd.xmu.edu.cn/detail.asp?serial=61884
dc.subject支持向量机
dc.subject变量选择
dc.subject强分层约束
dc.subjectSupport Vector Machine
dc.subjectVariable Selection
dc.subjectStrong Heredity Constraint
dc.title支持向量机中结构变量选择的研究
dc.title.alternativeResearch on Structure Variable Selection in Support Vector Machine
dc.typethesis
dc.date.replied2017-04-15
dc.description.note学位:经济学硕士
dc.description.note院系专业:经济学院_统计学
dc.description.note学号:15420141151996


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record