首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于Universum的多视角全局和局部结构风险最小化模型
引用本文:朱昌明,梅成就,周日贵,魏莱,章夏芬.基于Universum的多视角全局和局部结构风险最小化模型[J].上海海事大学学报,2018,39(3):97-104.
作者姓名:朱昌明  梅成就  周日贵  魏莱  章夏芬
作者单位:上海海事大学信息工程学院
基金项目:国家自然科学基金(61602296,61603245);上海市自然科学基金(16ZR1414500,16ZR1414400);上海市浦江人才计划(16PJ1403700)
摘    要:为克服传统多视角分类器无法充分最小化结构风险的不足,提出基于Universum的多视角全局和局部结构风险最小化模型。该模型采用Universum学习,利用有标签样本生成大量包含分类信息的无标签样本,从而增加分类器性能。这些信息有利于最小化结构风险。通过在Mfeat、Reuters和Corel等3个多视角数据集上的试验可以发现,该模型可以提高多视角分类器的性能,并可以更好地应用到多视角数据集的分类问题中。

关 键 词:Universum学习    多视角    结构风险
收稿时间:2017/6/9 0:00:00
修稿时间:2017/8/8 0:00:00

Universum-based multi-view global and local structural risk minimization model
Institution:Shanghai Maritime University, College of Information Engineering,Shanghai Maritime University, College of Information Engineering,Shanghai Maritime University, College of Information Engineering and Shanghai Maritime University, College of Information Engineering
Abstract:In order to overcome the disadvantage of traditional multi view classifiers that can not fully minimize structural risk, a Universum based multi view global and local structural risk minimization model is proposed. The model uses Universum learning, which uses labeled samples to generate a large number of unlabeled samples containing classification information so as to enhance the performances of classifiers. This information helps minimize structural risks. Experiments on three multi view data sets, i.e., Mfeat, Reuters and Corel, show that the model can improve the performance of multi view classifiers and can be better applied to the classification of multi view data sets.
Keywords:Universum learning  multi-view  structural risk
本文献已被 CNKI 等数据库收录!
点击此处可从《上海海事大学学报》浏览原始摘要信息
点击此处可从《上海海事大学学报》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号