Multi-view classification with cross-view must-link and cannot-link side information.pdfVIP

下载本文档

8
0
约4.79万字
约 10页
2017-03-16 发布于天津
举报
版权申诉

Multi-view classification with cross-view must-link and cannot-link side information.pdf

1、本文档共10页，可阅读全部内容。
2、原创力文档（book118）网站文档一经付费（服务费），不意味着购买了该文档的版权，仅供个人/单位学习、研究之用，不得用于商业用途，未经授权，严禁复制、发行、汇编、翻译或者网络传播等，侵权必究。
3、本站所有内容均由合作方或网友上传，本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺！文档内容仅供研究参考，付费前请自行鉴别。如您付费，意味着您自己接受本站规则且自行承担风险，本站不退款、不进行额外附加服务；查看《如何避免下载的几个坑》。如果您已付费下载过本站文档，您可以点击这里二次下载。
4、如文档侵犯商业秘密、侵犯著作权、侵犯人身权等，请点击“版权申诉”（推荐），也可以打举报电话：400-050-0827(电话支持时间：9:00-18:30)。
5、该文档为VIP文档，如果想要下载，成为VIP会员后，下载免费。
6、成为VIP后，下载本文档将扣除1次下载权益。下载后，不支持退款、换文档。如有疑问请联系我们。
7、成为VIP后，您将拥有八大权益，权益包括：VIP文档下载权益、阅读免打扰、文档格式转换、高级专利检索、专属身份标志、高级客服、多端互通、版权登记。
8、VIP文档为合作方或网友上传，每下载1次，网站将根据用户上传文档的质量评分、类型等，对文档贡献者给予高额补贴、流量扶持。如果你也想贡献VIP文档。上传文档

Multi-view classification with cross-view must-link and cannot-link side information.pdf

Knowledge-Based Systems 54 (2013) 137–146 Contents lists available at ScienceDirect Knowledge-Based Systems journal homepage: /locate/knosys Multi-view classi?cation with cross-view must-link and cannot-link side information Qiang Qian a, Songcan Chen a,?, Xudong Zhou b a Department of Computer Science and Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, P.R. China b Information Engineering College, Yangzhou University, Yangzhou 225009, P.R. China article info Article history: Received 22 October 2012 Received in revised form 1 September 2013 Accepted 5 September 2013 Available online 9 October 2013 Keywords: Classi?cation Multi-view learning Without correspondence Unpaired multi-view data Cross-view side information abstract Side information, like must-link (ML) and cannot-link (CL), has been widely used in single-view classi?cation tasks. However, so far such information has never been applied in multi-view classi?cation tasks. In many real world situations, data with multiple representations or views are frequently encountered, and most proposed algorithms for such learning situations require that all the multi-view data should be paired. Yet this requirement is dif?cult to satisfy in some settings and the multi-view data could be totally unpaired. In this paper, we propose an learning framework to design the multi-view classi?ers by only employing the weak side information of cross-view must-links (CvML) and cross-view cannotlinks (CvCL). The CvML and the CvCL generalize the traditional single-view must-link (SvML) and single-view cannot-link (SvCL), and to the best of our knowledge, are ?rst de?nitely introduced and applied into the multi-view classi?cation situations. Finally, we demonstrate the effectiveness of our method in our experiments. ó 2013 Elsevier B.V. All rights reserved. 1. Introduction Traditional learning only involves data with single view. However, in many real world circumstances, data with multiple natural fea