Article(id=1329419443567965113, tenantId=1146029695717560320, journalId=1324759914146529478, issueId=1329419378451394604, articleNumber=null, orderNo=null, doi=10.7544/issn1000-1239.202550353, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1748188800000, receivedDateStr=2025-05-26, revisedDate=1758643200000, revisedDateStr=2025-09-24, acceptedDate=1760112000000, acceptedDateStr=2025-10-11, onlineDate=1794887894765, onlineDateStr=2026-11-17, pubDate=1774972800000, pubDateStr=2026-04-01, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1794887894765, onlineIssueDateStr=2026-11-17, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1794887894765, creator=13701087609, updateTime=1794887894765, updator=13701087609, issue=Issue{id=1329419378451394604, tenantId=1146029695717560320, journalId=1324759914146529478, year='2026', volume='63', issue='4', pageStart='825', pageEnd='1093', issueExtLink='null', onlineDate='null', pubDate='1774972800000', pubDateStr='2026-04-01', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1794887879240, creator='13701087609', updateTime=1794887879240, updator='13701087609', preIssue=null, nextIssue=null, articleTotal=null, ext=null, issueFiles=null, downloadFileDto=null}, startPage=960, endPage=972, ext={EN=ArticleExt(id=1329419443777680314, articleId=1329419443567965113, tenantId=1146029695717560320, journalId=1324759914146529478, language=EN, title=Multi-Layer Contrastive Multi-View Clustering via Spectral Embedding Similarity, columnId=1325489891789910943, journalTitle=Journal of Computer Research and Development, columnName=Artificial Intelligence, runingTitle=null, highlight=null, articleAbstract=

Multi-view clustering, which learns consensus representations in an unsupervised manner to partition data samples into distinct categories, has gained widespread attention. In recent years, contrastive learning has demonstrated significant potential in multi-view clustering due to its powerful capability for cross-view feature alignment. However, existing contrastive multi-view clustering methods suffer from false negatives and false positives in positive-negative sample partitioning, limiting further improvements in clustering performance. To address these issues, we propose a multi-layer contrastive multi-view clustering via spectral embedding similarity (MCSES). By introducing a multi-layer contrastive learning mechanism grounded in spectral embedding similarity, MCSES effectively mitigates the performance bottleneck caused by inaccurate sample partitioning. Specifically, Laplacian matrices are first constructed based on the latent representations of each view, and their eigenvectors are utilized to generate spectral embedding similarity matrices. Subsequently, positive and negative sample pairs are partitioned according to the global prior information provided by these spectral embedding similarity matrices, alleviating the interference of sample partitioning errors on latent representation learning. Then, a cross-view contrastive loss is designed to strengthen inter-view consistency and align shared semantics. Finally, the latent representations from each view are adaptively fused through learnable weights to obtain a global multi-view representation, upon which a global contrastive loss is designed to emphasize multi-view consensus information. Extensive experiments on public multi-view datasets validate the effectiveness of the proposed method.

, authors=Xiaoyang Ren, Yuxiu Lin, Xiao Yu, Hui Liu*, authorsList=Xiaoyang Ren, Yuxiu Lin, Xiao Yu, Hui Liu, authorCompany=null, correspAuthors=Hui Liu, authorNote=

born in 2003. Master. Student member of CCF. His main research interests include multi-view clustering and data mining

born in 1996. PhD. Her main research interests include data mining and visualization, and multi-view learning

born in 1981. PhD, associate professor. Member of CCF. Her main research interests include machine learning and data mining

born in 1978. PhD, professor, PhD supervisor. Member of CCF. Her main research interests include machine learning, data mining, and image processing

, correspAuthorsNote=null, copyrightStatement=Copyright © 2026 Journal of Computer Research and Development. All rights reserved., copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=null, magXml=null, pdfUrl=null, pdf=null, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=null, mapNumber=null, fund=null), CN=ArticleExt(id=1329419445644145621, articleId=1329419443567965113, tenantId=1146029695717560320, journalId=1324759914146529478, language=CN, title=基于谱嵌入相似度的多层对比多视图聚类, columnId=1325489892104483745, journalTitle=计算机研究与发展, columnName=人工智能, runingTitle=null, highlight=null, articleAbstract=

多视图聚类通过将数据样本划分为不同类别的无监督方式学习共识表示,受到了广泛关注。近年来,对比学习凭借其强大的跨视图特征对齐能力,在多视图聚类中展现出了巨大的潜力。然而,现有对比多视图聚类方法存在正负样本划分的假阴性与假阳性问题,限制了聚类性能的进一步提升。为解决这些问题,提出了基于谱嵌入相似度的多层对比多视图聚类方法。通过引入基于谱嵌入相似度的多层对比学习机制,有效缓解了正负样本划分不准确带来的性能瓶颈。首先,基于各视图潜在表示构造拉普拉斯矩阵,并利用其特征向量生成谱嵌入相似度矩阵。其次,根据谱嵌入相似度矩阵提供的全局先验信息划分正负样本对,缓解了样本划分错误对潜在表示学习的干扰。然后,设计跨视图对比损失来强化一致性表征和共享语义对齐。最后,通过可学习权重自适应地融合各个视图的潜在表示,得到全局多视图表示,并在此基础上设计全局对比损失来强调多视图一致性信息。在公开多视图数据集上的大量实验验证了所提方法的有效性。

, authors=任晓阳, 林毓秀, 于晓, 刘慧*, authorsList=任晓阳, 林毓秀, 于晓, 刘慧, authorCompany=null, correspAuthors=刘慧, authorNote=

2003年生。硕士。CCF学生会员。主要研究方向为多视图聚类、数据挖掘

1996年生。博士。主要研究方向为数据挖掘与可视化、多视图学习

1981年生。博士,副教授。CCF会员。主要研究方向为机器学习、数据挖掘

1978年生。博士,教授,博士生导师。CCF会员。主要研究方向为机器学习、数据挖掘、图像处理

, correspAuthorsNote=
刘慧()
, copyrightStatement=版权所有 © 2026 计算机研究与发展编辑部, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=nZ20CN0OX/mdlzAVzD1y6g==, magXml=U6b4LZreYzjAvkadwwXLuA==, pdfUrl=null, pdf=F9ETo9ltoX3cJhwaoahUsg==, pdfFileSize=3263089, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=Mx2iUiEkCKe4a7plhcXenw==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=UJI6YmXS5VpPdC4wveJ1Sg==, mapNumber=null, fund=null)}, authors=[Author(id=1329419446193599453, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=sduferxy@163.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1329419446256514014, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, authorId=1329419446193599453, language=EN, stringName=Xiaoyang Ren, firstName=Xiaoyang, middleName=null, lastName=Ren, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=null, bio={"content":"

born in 2003. Master. Student member of CCF. His main research interests include multi-view clustering and data mining

"}, bioImg=null, bioContent=

born in 2003. Master. Student member of CCF. His main research interests include multi-view clustering and data mining

, aboutCorrespAuthor=null), CN=AuthorExt(id=1329419446348788704, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, authorId=1329419446193599453, language=CN, stringName=任晓阳, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=null, bio={"img":"nNBCpozdrCwuFRe1qglbcw==","content":"

2003年生。硕士。CCF学生会员。主要研究方向为多视图聚类、数据挖掘

"}, bioImg=nNBCpozdrCwuFRe1qglbcw==, bioContent=

2003年生。硕士。CCF学生会员。主要研究方向为多视图聚类、数据挖掘

, aboutCorrespAuthor=null)}, companyList=null), Author(id=1329419446415897570, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1329419446483006435, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, authorId=1329419446415897570, language=EN, stringName=Yuxiu Lin, firstName=Yuxiu, middleName=null, lastName=Lin, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=null, bio={"content":"

born in 1996. PhD. Her main research interests include data mining and visualization, and multi-view learning

"}, bioImg=null, bioContent=

born in 1996. PhD. Her main research interests include data mining and visualization, and multi-view learning

, aboutCorrespAuthor=null), CN=AuthorExt(id=1329419446541726692, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, authorId=1329419446415897570, language=CN, stringName=林毓秀, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=null, bio={"img":"aYk0bYBg8YfGjbfgSKyjmw==","content":"

1996年生。博士。主要研究方向为数据挖掘与可视化、多视图学习

"}, bioImg=aYk0bYBg8YfGjbfgSKyjmw==, bioContent=

1996年生。博士。主要研究方向为数据挖掘与可视化、多视图学习

, aboutCorrespAuthor=null)}, companyList=null), Author(id=1329419446634001383, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1329419446738858983, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, authorId=1329419446634001383, language=EN, stringName=Xiao Yu, firstName=Xiao, middleName=null, lastName=Yu, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=null, bio={"content":"

born in 1981. PhD, associate professor. Member of CCF. Her main research interests include machine learning and data mining

"}, bioImg=null, bioContent=

born in 1981. PhD, associate professor. Member of CCF. Her main research interests include machine learning and data mining

, aboutCorrespAuthor=null), CN=AuthorExt(id=1329419446852105192, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, authorId=1329419446634001383, language=CN, stringName=于晓, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=null, bio={"img":"Pm4H8XZA6MEOdID3Ne2+9Q==","content":"

1981年生。博士,副教授。CCF会员。主要研究方向为机器学习、数据挖掘

"}, bioImg=Pm4H8XZA6MEOdID3Ne2+9Q==, bioContent=

1981年生。博士,副教授。CCF会员。主要研究方向为机器学习、数据挖掘

, aboutCorrespAuthor=null)}, companyList=null), Author(id=1329419446927602666, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=liuh_lh@sdufe.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1329419447036654572, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, authorId=1329419446927602666, language=EN, stringName=Hui Liu, firstName=Hui, middleName=null, lastName=Liu, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=*, address=null, bio={"content":"

born in 1978. PhD, professor, PhD supervisor. Member of CCF. Her main research interests include machine learning, data mining, and image processing

"}, bioImg=null, bioContent=

born in 1978. PhD, professor, PhD supervisor. Member of CCF. Her main research interests include machine learning, data mining, and image processing

, aboutCorrespAuthor=null), CN=AuthorExt(id=1329419447158289389, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, authorId=1329419446927602666, language=CN, stringName=刘慧, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=*, address=null, bio={"img":"elpae/LTNXv2Y6brbqIYIg==","content":"

1978年生。博士,教授,博士生导师。CCF会员。主要研究方向为机器学习、数据挖掘、图像处理

"}, bioImg=elpae/LTNXv2Y6brbqIYIg==, bioContent=

1978年生。博士,教授,博士生导师。CCF会员。主要研究方向为机器学习、数据挖掘、图像处理

, aboutCorrespAuthor=null)}, companyList=null)], keywords=[Keyword(id=1329419447292507118, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=EN, orderNo=1, keyword=multi-view clustering), Keyword(id=1329419447372198895, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=EN, orderNo=2, keyword=contrastive learning), Keyword(id=1329419447464473584, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=EN, orderNo=3, keyword=spectral embedding), Keyword(id=1329419447602885618, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=EN, orderNo=4, keyword=Laplacian matrix), Keyword(id=1329419447686771699, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=EN, orderNo=5, keyword=weighted fusion), Keyword(id=1329419447795823605, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=CN, orderNo=1, keyword=多视图聚类), Keyword(id=1329419447934235638, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=CN, orderNo=2, keyword=对比学习), Keyword(id=1329419448026510327, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=CN, orderNo=3, keyword=谱嵌入), Keyword(id=1329419448089424888, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=CN, orderNo=4, keyword=拉普拉斯矩阵), Keyword(id=1329419448156533753, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=CN, orderNo=5, keyword=加权融合)], refs=[Reference(id=1329419450912190489, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=1, rfOrder=0, authorNames=null, journalName=null, refType=null, unstructuredReference=刘新旺. 多视图学习的基础理论和方法研究[J]. 中国基础科学, 2022, 24(3): 27−34, articleTitle=null, refAbstract=null), Reference(id=1329419450996076571, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=1, rfOrder=1, authorNames=null, journalName=null, refType=null, unstructuredReference=Liu Xinwang. Research on the fundamental theory and methods of multi-view learning[J]. China Basic Science, 2022, 24(3): 27−34 (in Chinese), articleTitle=null, refAbstract=null), Reference(id=1329419451063185436, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=2, rfOrder=2, authorNames=null, journalName=null, refType=null, unstructuredReference=赵玉涵, 陈松灿. 消除互补性争议的多视图聚类算法[J]. 计算机研究与发展, 2025, 62(5): 1216−1225, articleTitle=null, refAbstract=null), Reference(id=1329419451130294301, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=2, rfOrder=3, authorNames=null, journalName=null, refType=null, unstructuredReference=Zhao Yuhan, Chen Songcan. Multi-view clustering algorithm via eliminating the complementarity controversy[J]. Journal of Computer Research and Development, 2025, 62(5): 1216−1225 (in Chinese), articleTitle=null, refAbstract=null), Reference(id=1329419451239346207, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=3, rfOrder=4, authorNames=null, journalName=null, refType=null, unstructuredReference=Chao Guoqing, Sun Shiliang, Bi Jinbo. A survey on multiview clustering[J]. IEEE Transactions on Artificial Intelligence, 2021, 2(2): 146−168, articleTitle=null, refAbstract=null), Reference(id=1329419451298066464, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=4, rfOrder=5, authorNames=null, journalName=null, refType=null, unstructuredReference=张荣国, 曹俊辉, 胡静, 等. 基于非负正交矩阵分解的多视图聚类图像分割算法[J]. 模式识别与人工智能, 2023, 36(6): 556−571, articleTitle=null, refAbstract=null), Reference(id=1329419451360981025, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=4, rfOrder=6, authorNames=null, journalName=null, refType=null, unstructuredReference=Zhang Rongguo, Cao Junhui, Hu Jing, et al. Non-negative orthogonal matrix factorization based multi-view clustering image segmentation algorithm[J]. Pattern Recognition and Artificial Intelligence, 2023, 36(6): 556−571(in Chinese), articleTitle=null, refAbstract=null), Reference(id=1329419451428089891, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=5, rfOrder=7, authorNames=null, journalName=null, refType=null, unstructuredReference=Zhang Yilin, Chen Shuo, Mccoy R G, et al. Multi-view k-nearest neighbor graph contrastive learning on multi-modal biomedical data[C]//Proc of the 22nd Int Conf on Artificial Intelligence in Medicine. Berlin: Springer, 2024: 238−248, articleTitle=null, refAbstract=null), Reference(id=1329419451520364580, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=6, rfOrder=8, authorNames=null, journalName=null, refType=null, unstructuredReference=Yan Xiaoqiang, Gan Yingtao, Mao Yiqiao, et al. Live and learn: Continual action clustering with incremental views[C]//Proc of the 38th AAAI Conf on Artificial Intelligence. Palo Alto, CA: AAAI, 2024: 16264−16271, articleTitle=null, refAbstract=null), Reference(id=1329419451600056357, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=7, rfOrder=9, authorNames=null, journalName=null, refType=null, unstructuredReference=Wu Zhirong, Xiong Yuanjun, Yu S X, et al. Unsupervised feature learning via non-parametric instance discrimination[C]//Proc of the IEEE Conf on Computer Vision and Pattern Recognition. Piscataway, NJ: IEEE, 2018: 3733−3742, articleTitle=null, refAbstract=null), Reference(id=1329419451700719654, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=8, rfOrder=10, authorNames=null, journalName=null, refType=null, unstructuredReference=Bian Jintang, Xie Xiaohua, Lai Jianhuang, et al. Multi-view contrastive clustering via integrating graph aggregation and confidence enhancement[J]. Information Fusion, 2024, 108: 102393, articleTitle=null, refAbstract=null), Reference(id=1329419451801382951, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=9, rfOrder=11, authorNames=null, journalName=null, refType=null, unstructuredReference=Wang Haiyue, Wang Quan, Miao Qiguang, et al. Joint learning of data recovering and graph contrastive denoising for incomplete multi-view clustering[J]. Information Fusion, 2024, 104: 102155, articleTitle=null, refAbstract=null), Reference(id=1329419451864297512, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=10, rfOrder=12, authorNames=null, journalName=null, refType=null, unstructuredReference=Liu Jiyuan, Liu Xinwang, Yang Yuexiang, et al. Contrastive multi-view kernel learning[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023, 45(8): 9552−9566, articleTitle=null, refAbstract=null), Reference(id=1329419451935600681, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=11, rfOrder=13, authorNames=null, journalName=null, refType=null, unstructuredReference=Cui Jinrong, Li Yuting, Huang Han, et al. Dual contrast-driven deep multi-view clustering[J]. IEEE Transactions on Image Processing, 2024, 33: 4753−4764, articleTitle=null, refAbstract=null), Reference(id=1329419451994320938, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=12, rfOrder=14, authorNames=null, journalName=null, refType=null, unstructuredReference=Yang Xihong, Jin Jiaqi, Wang Siwei, et al. DealMVC: Dual contrastive calibration for multi-view clustering[C]//Proc of the 31st ACM Int Conf on Multimedia. New York: ACM, 2023: 337−346, articleTitle=null, refAbstract=null), Reference(id=1329419452053041195, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=13, rfOrder=15, authorNames=null, journalName=null, refType=null, unstructuredReference=Yan Weiqing, Zhang Yuanyang, Lv Chenlei, et al. GCFAgg: Global and cross-view feature aggregation for multi-view clustering[C]//Proc of the IEEE/CVF Conf on Computer Vision and Pattern Recognition. Piscataway, NJ: IEEE, 2023: 19863−19872, articleTitle=null, refAbstract=null), Reference(id=1329419452136927276, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=14, rfOrder=16, authorNames=null, journalName=null, refType=null, unstructuredReference=Yang Mouxing, Li Yunfan, Hu Peng, et al. Robust multi-view clustering with incomplete information[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 45(1): 1055−1069, articleTitle=null, refAbstract=null), Reference(id=1329419452266950702, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=15, rfOrder=17, authorNames=null, journalName=null, refType=null, unstructuredReference=Belkin M, Niyogi P. Laplacian eigenmaps and spectral techniques for embedding and clustering[C]//Proc of the 15th Int Conf on Neural Information Processing Systems: Natural and Synthetic. Cambridge, MA: MIT, 2001: 585−591, articleTitle=null, refAbstract=null), Reference(id=1329419452338253871, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=16, rfOrder=18, authorNames=null, journalName=null, refType=null, unstructuredReference=林毓秀, 刘慧, 于晓, 等. 面向子空间聚类的多视图统一表示学习网络[J]. 计算机研究与发展, 2025, 62(5): 1248−1261, articleTitle=null, refAbstract=null), Reference(id=1329419452430528560, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=16, rfOrder=19, authorNames=null, journalName=null, refType=null, unstructuredReference=Lin Yuxiu, Liu Hui, Yu Xiao, et al. A multi-view unified representation learning network for subspace clustering[J]. Journal of Computer Research and Development, 2025, 62(5): 1248−1261(in Chinese), articleTitle=null, refAbstract=null), Reference(id=1329419452497637425, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=17, rfOrder=20, authorNames=null, journalName=null, refType=null, unstructuredReference=Cai Xiaosha, Huang Dong, Zhang Guangyu, et al. Seeking commonness and inconsistencies: A jointly smoothed approach to multi-view subspace clustering[J]. Information Fusion, 2023, 91: 364−375, articleTitle=null, refAbstract=null), Reference(id=1329419452568940595, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=18, rfOrder=21, authorNames=null, journalName=null, refType=null, unstructuredReference=Ji Jintian, Feng Songhe. Anchor structure regularization induced multi-view subspace clustering via enhanced tensor rank minimization[C]//Proc of the IEEE Int Conf on Computer Vision. Piscataway, NJ: IEEE, 2023: 19343−19352, articleTitle=null, refAbstract=null), Reference(id=1329419452631855156, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=19, rfOrder=22, authorNames=null, journalName=null, refType=null, unstructuredReference=Liu Jialu, Wang Chi, Gao Jing, et al. Multi-view clustering via joint nonnegative matrix factorization[C]//Proc of the 2013 SIAM Int Conf on Data Mining. Philadelphia, PA: SIAM, 2013: 252−260, articleTitle=null, refAbstract=null), Reference(id=1329419452694769717, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=20, rfOrder=23, authorNames=null, journalName=null, refType=null, unstructuredReference=Wei Shaowei, Wang Jun, Yu Guoxian, et al. Multi-view multiple clusterings using deep matrix factorization[C]//Proc of the 34th AAAI Conf on Artificial Intelligence. Palo Alto, CA: AAAI, 2020: 6348−6355, articleTitle=null, refAbstract=null), Reference(id=1329419452774461494, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=21, rfOrder=24, authorNames=null, journalName=null, refType=null, unstructuredReference=Jing Peiguang, Su Yuting, Li Zhengnan, et al. Learning robust affinity graph representation for multi-view clustering[J]. Information Sciences, 2021, 544: 155−167, articleTitle=null, refAbstract=null), Reference(id=1329419452849958967, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=22, rfOrder=25, authorNames=null, journalName=null, refType=null, unstructuredReference=Pan Erlin, Kang Zhao. Multi-view contrastive graph clustering[C]//Proc of the 35th Int Conf on Neural Information Processing Systems. Cambridge, MA: MIT, 2021: 2148−2159, articleTitle=null, refAbstract=null), Reference(id=1329419453013536825, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=23, rfOrder=26, authorNames=null, journalName=null, refType=null, unstructuredReference=Li Lusi, He Haibo. Bipartite graph based multi-view clustering[J]. IEEE Transactions on Knowledge and Data Engineering, 2020, 34(7): 3111−3125, articleTitle=null, refAbstract=null), Reference(id=1329419453202280506, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=24, rfOrder=27, authorNames=null, journalName=null, refType=null, unstructuredReference=Fan Shaohua, Wang Xiao, Shi Chuan, et al. One2multi graph autoencoder for multi-view graph clustering[C]//Proc of the Web Conf 2020. New York: ACM, 2020: 3070−3076, articleTitle=null, refAbstract=null), Reference(id=1329419453281972283, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=25, rfOrder=28, authorNames=null, journalName=null, refType=null, unstructuredReference=Tang Chang, Liu Xinwang, Zhu Xinzhong, et al. CGD: Multi-view clustering via cross-view graph diffusion[C]//Proc of the 34th AAAI Conf on Artificial Intelligence. Palo Alto, CA: AAAI, 2020: 5924−5931, articleTitle=null, refAbstract=null), Reference(id=1329419453432967229, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=26, rfOrder=29, authorNames=null, journalName=null, refType=null, unstructuredReference=Xu Jie, Ren Yazhou, Tang Huayi, et al. Multi-VAE: Learning disentangled view-common and view-peculiar visual representations for multi-view clustering[C]//Proc of the IEEE Int Conf on Computer Vision. Piscataway, NJ: IEEE, 2021: 9234−9243, articleTitle=null, refAbstract=null), Reference(id=1329419453533630526, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=27, rfOrder=30, authorNames=null, journalName=null, refType=null, unstructuredReference=Dong Zhibin, Wang Siwei, Jin Jiaqi, et al. Cross-view topology based consistent and complementary information for deep multi-view clustering[C]//Proc of the IEEE Int Conf on Computer Vision. Piscataway, NJ: IEEE, 2023: 19440−19451, articleTitle=null, refAbstract=null), Reference(id=1329419453596545087, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=28, rfOrder=31, authorNames=null, journalName=null, refType=null, unstructuredReference=Wu Song, Zheng Yan, Ren Yazhou, et al. Self-weighted contrastive fusion for deep multi-view clustering[J]. IEEE Transactions on Multimedia, 2024, 26: 9150−9162, articleTitle=null, refAbstract=null), Reference(id=1329419453680431168, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=29, rfOrder=32, authorNames=null, journalName=null, refType=null, unstructuredReference=Luo Caixuan, Xu Jie, Ren Yazhou, et al. Simple contrastive multi-view clustering with data-level fusion[C]//Proc of the 33rd Int Joint Conf on Artificial Intelligence. San Francisco, CA: Morgan Kaufmann, 2024: 4697−4705, articleTitle=null, refAbstract=null), Reference(id=1329419453747540033, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=30, rfOrder=33, authorNames=null, journalName=null, refType=null, unstructuredReference=Zhang Qian, Zhang Lin, Song Ran, et al. Learning common semantics via optimal transport for contrastive multi-view clustering[J]. IEEE Transactions on Image Processing, 2024, 33: 4501−4515, articleTitle=null, refAbstract=null), Reference(id=1329419453839814722, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=31, rfOrder=34, authorNames=null, journalName=null, refType=null, unstructuredReference=Hinton G E, Salakhutdinov R R. Reducing the dimensionality of data with neural networks[J]. Science, 2006, 313(5786): 504−507, articleTitle=null, refAbstract=null), Reference(id=1329419453936283716, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=32, rfOrder=35, authorNames=null, journalName=null, refType=null, unstructuredReference=Coates A, Ng A, Lee H. An analysis of single-layer networks in unsupervised feature learning[C]//Proc of the 14th Int Conf on Artificial Intelligence and Statistics. Brookline, MA: JMLR W&CP, 2011: 215−223, articleTitle=null, refAbstract=null), Reference(id=1329419454015975493, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=33, rfOrder=36, authorNames=null, journalName=null, refType=null, unstructuredReference=Wen Jie, Zhang Zheng, Zhang Zhao, et al. Dimc-net: Deep incomplete multi-view clustering network[C]//Proc of the 28th ACM Int Conf on Multimedia. New York: ACM, 2020: 3753−3761, articleTitle=null, refAbstract=null), Reference(id=1329419454083084358, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=34, rfOrder=37, authorNames=null, journalName=null, refType=null, unstructuredReference=Lu Yiding, Lin Yijie, Yang Mouxing, et al. Decoupled contrastive multi-view clustering with high-order random walks[C]//Proc of the 38th AAAI Conf on Artificial Intelligence. Palo Alto, CA: AAAI, 2024: 14193−14201, articleTitle=null, refAbstract=null), Reference(id=1329419454150193223, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=35, rfOrder=38, authorNames=null, journalName=null, refType=null, unstructuredReference=Fang Siguo Huang Dong, Cai Xiaosha, et al. Efficient multi-view clustering via unified and discrete bipartite graph learning[J]. IEEE Transactions on Neural Networks and Learning Systems, 2023, 35(8): 11436−11447, articleTitle=null, refAbstract=null), Reference(id=1329419454221496392, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=36, rfOrder=39, authorNames=null, journalName=null, refType=null, unstructuredReference=Lin Yijie, Gou Yuanbiao, Liu Xiaotian, et al. Dual contrastive prediction for incomplete multi-view representation learning[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 45(4): 4447−4461, articleTitle=null, refAbstract=null), Reference(id=1329419454301188170, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=37, rfOrder=40, authorNames=null, journalName=null, refType=null, unstructuredReference=Peng Xi, Huang Zhenyu, Lv Jiancheng, et al. COMIC: Multi-view clustering without parameter selection[C]//Proc of the 36th Int Conf on Machine Learning. New York: ACM, 2019: 5092−5101, articleTitle=null, refAbstract=null), Reference(id=1329419454380879947, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=38, rfOrder=41, authorNames=null, journalName=null, refType=null, unstructuredReference=Kong Chen, Lin Dahua, Bansal M, et al. What are you talking about? text-to-image coreference[C]//Proc of the IEEE Conf on Computer Vision and Pattern Recognition. Piscataway, NJ: IEEE, 2014: 3558−3565, articleTitle=null, refAbstract=null), Reference(id=1329419454477348940, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=39, rfOrder=42, authorNames=null, journalName=null, refType=null, unstructuredReference=Kumar A, Rai P, Daume H. Co-regularized multi-view spectral clustering[C]//Proc of the 25th Int Conf on Neural Information Processing Systems. Cambridge, MA: MIT, 2011: 1413−1421, articleTitle=null, refAbstract=null), Reference(id=1329419454552846413, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=40, rfOrder=43, authorNames=null, journalName=null, refType=null, unstructuredReference=Tang Huayi, Liu Yong. Deep safe multi-view clustering: Reducing the risk of clustering performance degradation caused by view increase[C]//Proc of the IEEE/CVF Conf on Computer Vision and Pattern Recognition. Piscataway, NJ: IEEE, 2022: 202−211, articleTitle=null, refAbstract=null), Reference(id=1329419454661898318, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=41, rfOrder=44, authorNames=null, journalName=null, refType=null, unstructuredReference=Ren Yazhou, Pu Jingyu, Cui Chenhang, et al. Dynamic weighted graph fusion for deep multi-view clustering[C]//Proc of the 33rd Int Joint Conf on Artificial Intelligence. San Francisco, CA: Morgan Kaufmann, 2024: 4842−4850, articleTitle=null, refAbstract=null), Reference(id=1329419454737395792, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=42, rfOrder=45, authorNames=null, journalName=null, refType=null, unstructuredReference=Xu Jie, Tang Huayi, Ren Yazhou, et al. Multi-level feature learning for contrastive multi-view clustering[C]//Proc of the IEEE/CVF Conf on Computer Vision and Pattern Recognition. Piscataway, NJ: IEEE, 2022: 16051−16060, articleTitle=null, refAbstract=null)], funds=null, companyList=[AuthorCompany(id=1329419445979689942, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, xref=null, ext=[AuthorCompanyExt(id=1329419445992272855, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, companyId=1329419445979689942, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=School of Computing and Artificial Intelligence, Shandong University of Finance and Economics, Jinan 250014), AuthorCompanyExt(id=1329419446000661464, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, companyId=1329419445979689942, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=山东财经大学计算机与人工智能学院 济南 250014)]), AuthorCompany(id=1329419446109713369, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, xref=null, ext=[AuthorCompanyExt(id=1329419446122296282, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, companyId=1329419446109713369, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Shandong Key Laboratory of Lightweight Intelligent Computing and Visualization for Digital Economy (Shandong University of Finance and Economics), Jinan 250014), AuthorCompanyExt(id=1329419446130684891, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, companyId=1329419446109713369, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=山东省数字经济轻量智算与可视化重点实验室(山东财经大学) 济南 250014)])], figs=[ArticleFig(id=1329419448315917307, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=EN, label=Figure 1, caption=Architecture of MCSES, figureFileSmall=VIS3j55aqbv+X0cVPNLa3A==, figureFileBig=Mx2iUiEkCKe4a7plhcXenw==, tableContent=null), ArticleFig(id=1329419448383026172, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=CN, label=图1, caption=MCSES的架构, figureFileSmall=VIS3j55aqbv+X0cVPNLa3A==, figureFileBig=Mx2iUiEkCKe4a7plhcXenw==, tableContent=null), ArticleFig(id=1329419448550798333, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=EN, label=Figure 2, caption=Visualization of clustering, figureFileSmall=B/Ar2W+ETSCrqQmyWNnw9A==, figureFileBig=EGpl49NbTRP6c8lakudNDw==, tableContent=null), ArticleFig(id=1329419448630490110, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=CN, label=图2, caption=聚类可视化, figureFileSmall=B/Ar2W+ETSCrqQmyWNnw9A==, figureFileBig=EGpl49NbTRP6c8lakudNDw==, tableContent=null), ArticleFig(id=1329419448705987583, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=EN, label=Figure 3, caption=Sensitivity analysis of $ \alpha $ and $\beta $, figureFileSmall=qPNYndVbjv/jc9ELo8j3SQ==, figureFileBig=KFh9A7iLGvlq+9+k+Pbv4g==, tableContent=null), ArticleFig(id=1329419448785678336, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=CN, label=图3, caption= $ \alpha $和$\beta $的敏感性分析, figureFileSmall=qPNYndVbjv/jc9ELo8j3SQ==, figureFileBig=KFh9A7iLGvlq+9+k+Pbv4g==, tableContent=null), ArticleFig(id=1329419448932478977, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=EN, label=Figure 4, caption=Sensitivity analysis of $\tau $ and $f$, figureFileSmall=H8cye54IJ/sraOEdSAmPTQ==, figureFileBig=+7o6O2kOJzgPkakYUJFrag==, tableContent=null), ArticleFig(id=1329419449070891010, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=CN, label=图4, caption= $\tau $和$f$的敏感性分析, figureFileSmall=H8cye54IJ/sraOEdSAmPTQ==, figureFileBig=+7o6O2kOJzgPkakYUJFrag==, tableContent=null), ArticleFig(id=1329419449196720132, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=EN, label=Figure 5, caption=Effect of the weighted fusion module on clustering performance, figureFileSmall=bNf+QWN9JjhG7VnO6SfMRg==, figureFileBig=kmpGISBDVaiB16OSF/1ikA==, tableContent=null), ArticleFig(id=1329419449272217605, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=CN, label=图5, caption=加权融合模块对聚类性能的影响, figureFileSmall=bNf+QWN9JjhG7VnO6SfMRg==, figureFileBig=kmpGISBDVaiB16OSF/1ikA==, tableContent=null), ArticleFig(id=1329419449339326470, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=EN, label=Figure 6, caption=Convergence experiment, figureFileSmall=IsOqeS4Umn6zzY3nkkEcaw==, figureFileBig=EIO1TZEI2e3skhnWwOn+Bg==, tableContent=null), ArticleFig(id=1329419449398046727, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=CN, label=图6, caption=收敛性实验, figureFileSmall=IsOqeS4Umn6zzY3nkkEcaw==, figureFileBig=EIO1TZEI2e3skhnWwOn+Bg==, tableContent=null), ArticleFig(id=1329419449590984712, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=EN, label=Table 1, caption=

Main Notations

, figureFileSmall=null, figureFileBig=null, tableContent=
符号定义
$X$多视图数据集
$V$视图数量
$N$样本数量
$ {\boldsymbol {x}}_i^v $第$ v $个视图中第$ i $个样本的数据
$ {\boldsymbol {\hat x}}_i^v $第$ v $个视图中第$ i $个样本的重构数据
${\boldsymbol {z}}_i^v$第$ v $个视图中第$ i $个样本的特征表示
${{\boldsymbol {z}}_i}$第$ i $个样本的全局特征表示
${w_v}$第$ v $个视图的加权系数
${P_i}$全局特征中第$ i $个样本的正样本集合
${N_i}$全局特征中第$ i $个样本的负样本集合
$P_i^{\text{c}}$跨视图特征中第$ i $个样本的正样本集合
$N_i^{\text{c}}$跨视图特征中第$ i $个样本的负样本集合
$f$阈值
${\boldsymbol {M}}$全局特征谱嵌入相似度矩阵
$ {{\boldsymbol {M}}^{\text{c}}} $跨视图特征谱嵌入相似度矩阵
$\tau $温度系数
${\zeta _i}$样本$ i $在所有视图中的索引的集合
), ArticleFig(id=1329419449716813834, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=CN, label=表1, caption=

主要符号

, figureFileSmall=null, figureFileBig=null, tableContent=
符号定义
$X$多视图数据集
$V$视图数量
$N$样本数量
$ {\boldsymbol {x}}_i^v $第$ v $个视图中第$ i $个样本的数据
$ {\boldsymbol {\hat x}}_i^v $第$ v $个视图中第$ i $个样本的重构数据
${\boldsymbol {z}}_i^v$第$ v $个视图中第$ i $个样本的特征表示
${{\boldsymbol {z}}_i}$第$ i $个样本的全局特征表示
${w_v}$第$ v $个视图的加权系数
${P_i}$全局特征中第$ i $个样本的正样本集合
${N_i}$全局特征中第$ i $个样本的负样本集合
$P_i^{\text{c}}$跨视图特征中第$ i $个样本的正样本集合
$N_i^{\text{c}}$跨视图特征中第$ i $个样本的负样本集合
$f$阈值
${\boldsymbol {M}}$全局特征谱嵌入相似度矩阵
$ {{\boldsymbol {M}}^{\text{c}}} $跨视图特征谱嵌入相似度矩阵
$\tau $温度系数
${\zeta _i}$样本$ i $在所有视图中的索引的集合
), ArticleFig(id=1329419449783922699, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=EN, label=Table 2, caption=

Details of Dataset

, figureFileSmall=null, figureFileBig=null, tableContent=
数据集视图数类别数样本数
Cora472708
Digit-Product21030000
LandUse-213212100
MNIST-USPS2105000
RGB-D2131449
Synthetic3d33600
), ArticleFig(id=1329419449897168908, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=CN, label=表2, caption=

数据集详细信息

, figureFileSmall=null, figureFileBig=null, tableContent=
数据集视图数类别数样本数
Cora472708
Digit-Product21030000
LandUse-213212100
MNIST-USPS2105000
RGB-D2131449
Synthetic3d33600
), ArticleFig(id=1329419449985249293, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=EN, label=Table 3, caption=

Clustering Results on Cora, RGB-D and LandUse-21 Datasets

, figureFileSmall=null, figureFileBig=null, tableContent=
方法CoraRGB-DLandUse-21
ACCNMIPURARIACCNMIPURARIACCNMIPURARI
DSMVC0.2970.0970.3750.0630.4400.3880.5690.2610.2820.3360.3160.141
MFLVC0.3840.1980.4520.1410.4190.2320.4340.1890.2250.2570.2480.089
GCFAgg0.2810.1070.3770.0660.3330.3310.4990.1760.2430.3830.2920.133
DealMVC0.6110.3970.6110.3520.5190.3540.5230.3160.1830.1780.1960.053
SCM0.5760.4020.6290.3250.4170.3460.5340.2120.2820.3110.2960.131
DFMVC0.5600.3940.6190.2920.3870.3080.4890.2150.1710.2230.2050.047
CSOT0.3690.2450.4810.1670.4250.2450.4260.2180.2240.2510.2450.087
SCMVC0.5790.3670.5830.2940.4200.3140.5290.2520.2650.3060.2890.119
DCMVC0.6660.4960.7040.4290.5110.4110.6160.3210.2970.3490.3250.154
MCSES(本文)0.7060.4830.7100.4770.5580.4220.5960.3020.3090.3700.3520.147
), ArticleFig(id=1329419450098495503, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=CN, label=表3, caption=

Cora,RGB-D,LandUse-21数据集上的聚类结果

, figureFileSmall=null, figureFileBig=null, tableContent=
方法CoraRGB-DLandUse-21
ACCNMIPURARIACCNMIPURARIACCNMIPURARI
DSMVC0.2970.0970.3750.0630.4400.3880.5690.2610.2820.3360.3160.141
MFLVC0.3840.1980.4520.1410.4190.2320.4340.1890.2250.2570.2480.089
GCFAgg0.2810.1070.3770.0660.3330.3310.4990.1760.2430.3830.2920.133
DealMVC0.6110.3970.6110.3520.5190.3540.5230.3160.1830.1780.1960.053
SCM0.5760.4020.6290.3250.4170.3460.5340.2120.2820.3110.2960.131
DFMVC0.5600.3940.6190.2920.3870.3080.4890.2150.1710.2230.2050.047
CSOT0.3690.2450.4810.1670.4250.2450.4260.2180.2240.2510.2450.087
SCMVC0.5790.3670.5830.2940.4200.3140.5290.2520.2650.3060.2890.119
DCMVC0.6660.4960.7040.4290.5110.4110.6160.3210.2970.3490.3250.154
MCSES(本文)0.7060.4830.7100.4770.5580.4220.5960.3020.3090.3700.3520.147
), ArticleFig(id=1329419450224324624, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=EN, label=Table 4, caption=

Clustering Results on MNIST-USPS, Digit-Product and Synthetic3d Datasets

, figureFileSmall=null, figureFileBig=null, tableContent=
方法MNIST-USPSDigit-ProductSynthetic3d
ACCNMIPURARIACCNMIPURARIACCNMIPURARI
DSMVC0.8270.8300.8270.7370.8080.7320.8080.6420.9580.8350.9580.879
MFLVC0.9950.9870.9950.9900.9870.9690.9870.9720.9730.8820.9730.921
GCFAgg0.9960.9880.9960.9910.9880.9700.9880.9750.9730.8850.9730.922
DealMVC0.9190.9160.9190.8640.9650.9450.9650.9320.8230.6220.8230.610
SCM0.9890.9700.9890.9770.9640.9160.9640.9230.9730.8850.9730.922
DFMVC0.9960.9890.9960.9920.9610.9220.9610.9180.9520.8220.9520.863
CSOT0.9950.9850.9950.9880.9940.9840.9940.9880.9750.8900.9750.927
SCMVC0.9970.9920.9970.9940.9940.9840.9940.9880.9730.8820.9730.922
DCMVC0.9900.9730.9900.9790.9860.9620.9860.9700.9900.9510.9900.970
MCSES(本文)0.9970.9920.9970.9940.9960.9870.9960.9910.9920.9570.9920.975
), ArticleFig(id=1329419450400485393, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=CN, label=表4, caption=

MNIST-USPS,Digit-Product,Synthetic3d数据集上的聚类结果

, figureFileSmall=null, figureFileBig=null, tableContent=
方法MNIST-USPSDigit-ProductSynthetic3d
ACCNMIPURARIACCNMIPURARIACCNMIPURARI
DSMVC0.8270.8300.8270.7370.8080.7320.8080.6420.9580.8350.9580.879
MFLVC0.9950.9870.9950.9900.9870.9690.9870.9720.9730.8820.9730.921
GCFAgg0.9960.9880.9960.9910.9880.9700.9880.9750.9730.8850.9730.922
DealMVC0.9190.9160.9190.8640.9650.9450.9650.9320.8230.6220.8230.610
SCM0.9890.9700.9890.9770.9640.9160.9640.9230.9730.8850.9730.922
DFMVC0.9960.9890.9960.9920.9610.9220.9610.9180.9520.8220.9520.863
CSOT0.9950.9850.9950.9880.9940.9840.9940.9880.9750.8900.9750.927
SCMVC0.9970.9920.9970.9940.9940.9840.9940.9880.9730.8820.9730.922
DCMVC0.9900.9730.9900.9790.9860.9620.9860.9700.9900.9510.9900.970
MCSES(本文)0.9970.9920.9970.9940.9960.9870.9960.9910.9920.9570.9920.975
), ArticleFig(id=1329419450488565779, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=EN, label=Table 5, caption=

Ablation Study

, figureFileSmall=null, figureFileBig=null, tableContent=
方案损失项CoraLandUse-21
$ {\mathcal{L}_{\text{r}}} $$ {\mathcal{L}_{{\text{con}}}} $$ {\mathcal{L}_{{\text{inter}}}} $$ {\mathcal{L}_{{\text{intra}}}} $ACCNMIPURARIACCNMIPURARI
A√0.5330.3450.6090.2640.2360.3120.2880.099
B√√0.6760.4630.6880.4440.2460.3010.2830.107
C√√0.6950.4800.7020.4630.2460.3020.2840.107
D√√0.5210.3340.5990.2480.3040.3490.3450.141
E√√√0.7060.4830.7100.4770.3090.3700.3520.147
), ArticleFig(id=1329419450580840469, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=CN, label=表5, caption=

消融实验

, figureFileSmall=null, figureFileBig=null, tableContent=
方案损失项CoraLandUse-21
$ {\mathcal{L}_{\text{r}}} $$ {\mathcal{L}_{{\text{con}}}} $$ {\mathcal{L}_{{\text{inter}}}} $$ {\mathcal{L}_{{\text{intra}}}} $ACCNMIPURARIACCNMIPURARI
A√0.5330.3450.6090.2640.2360.3120.2880.099
B√√0.6760.4630.6880.4440.2460.3010.2830.107
C√√0.6950.4800.7020.4630.2460.3020.2840.107
D√√0.5210.3340.5990.2480.3040.3490.3450.141
E√√√0.7060.4830.7100.4770.3090.3700.3520.147
), ArticleFig(id=1329419450664726551, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=EN, label=Table 6, caption=

Performance Evaluation Under Missing Views

, figureFileSmall=null, figureFileBig=null, tableContent=
缺失率/%MNIST-USPSSynthetic3dLandUse-21
ACCNMIPURARIACCNMIPURARIACCNMIPURARI
300.9300.8790.9300.8480.7150.4770.7150.4260.2160.2600.2510.090
200.9400.8980.9400.8620.7730.4190.7730.4270.2470.2840.2810.099
100.9710.9420.9710.9350.8800.6330.8800.6670.2770.3120.3110.118
00.9970.9920.9970.9940.9920.9570.9920.9750.3090.3700.3520.147
), ArticleFig(id=1329419450786361368, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, language=CN, label=表6, caption=

缺失视图下的性能评估

, figureFileSmall=null, figureFileBig=null, tableContent=
缺失率/%MNIST-USPSSynthetic3dLandUse-21
ACCNMIPURARIACCNMIPURARIACCNMIPURARI
300.9300.8790.9300.8480.7150.4770.7150.4260.2160.2600.2510.090
200.9400.8980.9400.8620.7730.4190.7730.4270.2470.2840.2810.099
100.9710.9420.9710.9350.8800.6330.8800.6670.2770.3120.3110.118
00.9970.9920.9970.9940.9920.9570.9920.9750.3090.3700.3520.147
)], attaches=null, journal=Journal(id=1324759606947315908, delFlag=0, nameCn=计算机研究与发展, nameEn=Journal of Computer Research and Development, nameHistory1=null, nameHistory2=null, issn=1000-1239, eissn=null, cn=11-1777/TP, coden=null, periodic=0, language=CN, oaType=null, ccby=null, superviseOffice=null, ownerOffice=null, pubOffice=null, editorOffice=null, officeType=null, aims=null, clcCode=null, officeProv=null, officeCity=null, officeAddr=null, officeZip=null, officeEmail=null, officePhone=null, editDirector=null, officeDirector=null, officeDirectorPhone=null, officeStaffNum=null, officeEmpNum=null, coverPicUrl=k2l8zgqwstuxxevFX2qxVA==, journalPrice=null, startedYear=null, abbrevIsoEn=Journal of Computer Research and Development, journalRemark=null, publicationField=null, createdTime=1793776903141, updatedTime=1793850698612, createdBy=18614031015, updatedBy=13701087609, firstLetterCn=J, firstLetterEn=J, subjectCode=Electronic Information, subjectName=null, subjectCodeEn=Electronic Information, subjectNameEn=null, picCn=k2l8zgqwstuxxevFX2qxVA==, picEn=r6k0Ry+RD+c/VXj5QarmDw==, jcr=null, cjcr=null, exts=[JournalExt(id=1325069127710286688, language=CN, name=计算机研究与发展, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1793850698640, updatedTime=1793850698640, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://crad.ict.ac.cn/journalx/authorLogOn.action, submissionEditorUrl=https://crad.ict.ac.cn/journalx/editorLogOn.action, submissionReviewUrl=https://crad.ict.ac.cn/journalx/expertLogOn.action, submissionCeEditorUrl=https://crad.ict.ac.cn/journalx/editorInChiefLogOn.action, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1325069127777395553, language=EN, name=Journal of Computer Research and Development, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1793850698656, updatedTime=1793850698656, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://crad.ict.ac.cn/journalx/authorLogOn.action, submissionEditorUrl=https://crad.ict.ac.cn/journalx/editorLogOn.action, submissionReviewUrl=https://crad.ict.ac.cn/journalx/expertLogOn.action, submissionCeEditorUrl=https://crad.ict.ac.cn/journalx/editorInChiefLogOn.action, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1324759914146529478, websiteList=[Website(id=1325070300139892771, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1324759914146529478, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN, language=CN, createTime=1793850978169, createBy=18614031015, updateTime=1793851040259, updateBy=18614031015, name=计算机研究与发展-中文, tplId=1146099689490845704, title=计算机研究与发展, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1325070893751349360, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=articleTextType, value=kx, createTime=1793851119697, updateTime=1793851119697, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893730377837, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=banner, value=null, createTime=1793851119692, updateTime=1793851119692, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893772320883, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=grayFlag, value=0, createTime=1793851119702, updateTime=1793851119702, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893721989228, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=logo, value=https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN/file/pic?fileId=U27tDsy717YF7XSPA+Vwdw==, createTime=1793851119690, updateTime=1793851119690, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893784903797, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=minRunFlag, value=0, createTime=1793851119705, updateTime=1793851119705, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893742960751, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN/file/pic, createTime=1793851119695, updateTime=1793851119695, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893780709492, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=silenceFlag, value=0, createTime=1793851119704, updateTime=1793851119704, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893738766446, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1793851119694, updateTime=1793851119694, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893755543665, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=themeColor, value=null, createTime=1793851119698, updateTime=1793851119698, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893763932274, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=themeStyle, value=null, createTime=1793851119700, updateTime=1793851119700, creator=18614031015, updator=18614031015)]), Website(id=1325070300580294698, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1324759914146529478, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/jsjyjyfz/EN, language=EN, createTime=1793850978274, createBy=18614031015, updateTime=1793851036638, updateBy=18614031015, name=计算机研究与发展-英文, tplId=1146101810881728533, title=Journal of Computer Research and Development, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1325070925288321148, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=articleTextType, value=kx, createTime=1793851127216, updateTime=1793851127216, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925263155321, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=banner, value=null, createTime=1793851127210, updateTime=1793851127210, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925309292671, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=grayFlag, value=0, createTime=1793851127221, updateTime=1793851127221, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925254766712, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=logo, value=https://castjournals.cast.org.cn/joweb/jsjyjyfz/EN/file/pic?fileId=U27tDsy717YF7XSPA+Vwdw==, createTime=1793851127208, updateTime=1793851127208, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925326069889, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=minRunFlag, value=0, createTime=1793851127225, updateTime=1793851127225, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925279932539, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/jsjyjyfz/EN/file/pic, createTime=1793851127214, updateTime=1793851127214, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925317681280, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=silenceFlag, value=0, createTime=1793851127223, updateTime=1793851127223, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925271543930, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1793851127212, updateTime=1793851127212, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925292515453, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=themeColor, value=null, createTime=1793851127217, updateTime=1793851127217, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925300904062, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=themeStyle, value=null, createTime=1793851127219, updateTime=1793851127219, creator=18614031015, updator=18614031015)])], journalTitle=计算机研究与发展, weixinUrl=null, journalUrl=https://crad.ict.ac.cn/, iacademicId=null, status=1, seqNo=null, journalTitleEn=Journal of Computer Research and Development, journalPhotoCn=k2l8zgqwstuxxevFX2qxVA==, journalPhotoEn=r6k0Ry+RD+c/VXj5QarmDw==, journalFirstLetter=J, journalRecommend=null, journalNew=null, journalCollection=null, jcrJf=null, cjcrJf=null, jcrJfStr=null, cjcrJfStr=null, submissionFirstDecision=null, sciSubjectClassification=null, casSubjectClassification=null, citeScore=null, totalCitationFrequency=null, icpCode=null, psCode=null, advertisingLicenseCode=null, copyrightInformation=null, country=null, option=, provinceCode=null, provinceName=null, collectFlag=false, interPubPlatform=, interPubPlatformUrl=null), detailUrlCn=https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN/10.7544/issn1000-1239.202550353, detailUrlEn=https://castjournals.cast.org.cn/joweb/jsjyjyfz/EN/10.7544/issn1000-1239.202550353, pdfUrlCn=https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN/PDF/10.7544/issn1000-1239.202550353, pdfUrlEn=https://castjournals.cast.org.cn/joweb/jsjyjyfz/EN/PDF/10.7544/issn1000-1239.202550353, aliStartDate=0, aliEndDate=0, collectionFlag=false, citedCount=null, citedUrl=null, previewStatus=0, delFlag=0, hasFullText=1, orderTime=1774972800000, fullTextJson=null, articleText=null, reference=null)
收藏切换
基于谱嵌入相似度的多层对比多视图聚类
任晓阳 , 林毓秀 , 于晓 , 刘慧 *
计算机研究与发展 | 人工智能 2026,63(4): 960-972
收起
收藏切换
计算机研究与发展 |人工智能 2026 , 63 (4) : 960 -972
基于谱嵌入相似度的多层对比多视图聚类
全屏
[Author(id=1329419446193599453, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=sduferxy@163.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1329419446256514014, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, authorId=1329419446193599453, language=EN, stringName=Xiaoyang Ren, firstName=Xiaoyang, middleName=null, lastName=Ren, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=null, bio={"content":"

born in 2003. Master. Student member of CCF. His main research interests include multi-view clustering and data mining

"}, bioImg=null, bioContent=

born in 2003. Master. Student member of CCF. His main research interests include multi-view clustering and data mining

, aboutCorrespAuthor=null), CN=AuthorExt(id=1329419446348788704, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, authorId=1329419446193599453, language=CN, stringName=任晓阳, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=null, bio={"img":"nNBCpozdrCwuFRe1qglbcw==","content":"

2003年生。硕士。CCF学生会员。主要研究方向为多视图聚类、数据挖掘

"}, bioImg=nNBCpozdrCwuFRe1qglbcw==, bioContent=

2003年生。硕士。CCF学生会员。主要研究方向为多视图聚类、数据挖掘

, aboutCorrespAuthor=null)}, companyList=null), Author(id=1329419446415897570, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1329419446483006435, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, authorId=1329419446415897570, language=EN, stringName=Yuxiu Lin, firstName=Yuxiu, middleName=null, lastName=Lin, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=null, bio={"content":"

born in 1996. PhD. Her main research interests include data mining and visualization, and multi-view learning

"}, bioImg=null, bioContent=

born in 1996. PhD. Her main research interests include data mining and visualization, and multi-view learning

, aboutCorrespAuthor=null), CN=AuthorExt(id=1329419446541726692, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, authorId=1329419446415897570, language=CN, stringName=林毓秀, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=null, bio={"img":"aYk0bYBg8YfGjbfgSKyjmw==","content":"

1996年生。博士。主要研究方向为数据挖掘与可视化、多视图学习

"}, bioImg=aYk0bYBg8YfGjbfgSKyjmw==, bioContent=

1996年生。博士。主要研究方向为数据挖掘与可视化、多视图学习

, aboutCorrespAuthor=null)}, companyList=null), Author(id=1329419446634001383, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1329419446738858983, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, authorId=1329419446634001383, language=EN, stringName=Xiao Yu, firstName=Xiao, middleName=null, lastName=Yu, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=null, bio={"content":"

born in 1981. PhD, associate professor. Member of CCF. Her main research interests include machine learning and data mining

"}, bioImg=null, bioContent=

born in 1981. PhD, associate professor. Member of CCF. Her main research interests include machine learning and data mining

, aboutCorrespAuthor=null), CN=AuthorExt(id=1329419446852105192, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, authorId=1329419446634001383, language=CN, stringName=于晓, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=null, bio={"img":"Pm4H8XZA6MEOdID3Ne2+9Q==","content":"

1981年生。博士,副教授。CCF会员。主要研究方向为机器学习、数据挖掘

"}, bioImg=Pm4H8XZA6MEOdID3Ne2+9Q==, bioContent=

1981年生。博士,副教授。CCF会员。主要研究方向为机器学习、数据挖掘

, aboutCorrespAuthor=null)}, companyList=null), Author(id=1329419446927602666, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=liuh_lh@sdufe.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1329419447036654572, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, authorId=1329419446927602666, language=EN, stringName=Hui Liu, firstName=Hui, middleName=null, lastName=Liu, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=*, address=null, bio={"content":"

born in 1978. PhD, professor, PhD supervisor. Member of CCF. Her main research interests include machine learning, data mining, and image processing

"}, bioImg=null, bioContent=

born in 1978. PhD, professor, PhD supervisor. Member of CCF. Her main research interests include machine learning, data mining, and image processing

, aboutCorrespAuthor=null), CN=AuthorExt(id=1329419447158289389, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329419443567965113, authorId=1329419446927602666, language=CN, stringName=刘慧, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=*, address=null, bio={"img":"elpae/LTNXv2Y6brbqIYIg==","content":"

1978年生。博士,教授,博士生导师。CCF会员。主要研究方向为机器学习、数据挖掘、图像处理

"}, bioImg=elpae/LTNXv2Y6brbqIYIg==, bioContent=

1978年生。博士,教授,博士生导师。CCF会员。主要研究方向为机器学习、数据挖掘、图像处理

, aboutCorrespAuthor=null)}, companyList=null)]
任晓阳 , 林毓秀, 于晓, 刘慧*
作者信息
  • 山东财经大学计算机与人工智能学院 济南 250014
  • 山东省数字经济轻量智算与可视化重点实验室(山东财经大学) 济南 250014
通讯作者:
作者简介:

2003年生。硕士。CCF学生会员。主要研究方向为多视图聚类、数据挖掘

1996年生。博士。主要研究方向为数据挖掘与可视化、多视图学习

1981年生。博士,副教授。CCF会员。主要研究方向为机器学习、数据挖掘

1978年生。博士,教授,博士生导师。CCF会员。主要研究方向为机器学习、数据挖掘、图像处理

Multi-Layer Contrastive Multi-View Clustering via Spectral Embedding Similarity
Xiaoyang Ren , Yuxiu Lin, Xiao Yu, Hui Liu*
Affiliations
  • School of Computing and Artificial Intelligence, Shandong University of Finance and Economics, Jinan 250014
  • Shandong Key Laboratory of Lightweight Intelligent Computing and Visualization for Digital Economy (Shandong University of Finance and Economics), Jinan 250014
  • born in 2003. Master. Student member of CCF. His main research interests include multi-view clustering and data mining

    born in 1996. PhD. Her main research interests include data mining and visualization, and multi-view learning

    born in 1981. PhD, associate professor. Member of CCF. Her main research interests include machine learning and data mining

    born in 1978. PhD, professor, PhD supervisor. Member of CCF. Her main research interests include machine learning, data mining, and image processing

出版时间: 2026-04-01 doi: 10.7544/issn1000-1239.202550353
文章导航
收藏切换

多视图聚类通过将数据样本划分为不同类别的无监督方式学习共识表示,受到了广泛关注。近年来,对比学习凭借其强大的跨视图特征对齐能力,在多视图聚类中展现出了巨大的潜力。然而,现有对比多视图聚类方法存在正负样本划分的假阴性与假阳性问题,限制了聚类性能的进一步提升。为解决这些问题,提出了基于谱嵌入相似度的多层对比多视图聚类方法。通过引入基于谱嵌入相似度的多层对比学习机制,有效缓解了正负样本划分不准确带来的性能瓶颈。首先,基于各视图潜在表示构造拉普拉斯矩阵,并利用其特征向量生成谱嵌入相似度矩阵。其次,根据谱嵌入相似度矩阵提供的全局先验信息划分正负样本对,缓解了样本划分错误对潜在表示学习的干扰。然后,设计跨视图对比损失来强化一致性表征和共享语义对齐。最后,通过可学习权重自适应地融合各个视图的潜在表示,得到全局多视图表示,并在此基础上设计全局对比损失来强调多视图一致性信息。在公开多视图数据集上的大量实验验证了所提方法的有效性。

多视图聚类  /  对比学习  /  谱嵌入  /  拉普拉斯矩阵  /  加权融合

Multi-view clustering, which learns consensus representations in an unsupervised manner to partition data samples into distinct categories, has gained widespread attention. In recent years, contrastive learning has demonstrated significant potential in multi-view clustering due to its powerful capability for cross-view feature alignment. However, existing contrastive multi-view clustering methods suffer from false negatives and false positives in positive-negative sample partitioning, limiting further improvements in clustering performance. To address these issues, we propose a multi-layer contrastive multi-view clustering via spectral embedding similarity (MCSES). By introducing a multi-layer contrastive learning mechanism grounded in spectral embedding similarity, MCSES effectively mitigates the performance bottleneck caused by inaccurate sample partitioning. Specifically, Laplacian matrices are first constructed based on the latent representations of each view, and their eigenvectors are utilized to generate spectral embedding similarity matrices. Subsequently, positive and negative sample pairs are partitioned according to the global prior information provided by these spectral embedding similarity matrices, alleviating the interference of sample partitioning errors on latent representation learning. Then, a cross-view contrastive loss is designed to strengthen inter-view consistency and align shared semantics. Finally, the latent representations from each view are adaptively fused through learnable weights to obtain a global multi-view representation, upon which a global contrastive loss is designed to emphasize multi-view consensus information. Extensive experiments on public multi-view datasets validate the effectiveness of the proposed method.

multi-view clustering  /  contrastive learning  /  spectral embedding  /  Laplacian matrix  /  weighted fusion
任晓阳, 林毓秀, 于晓, 刘慧. 基于谱嵌入相似度的多层对比多视图聚类. 计算机研究与发展, 2026 , 63 (4) : 960 -972 . DOI: 10.7544/issn1000-1239.202550353
Xiaoyang Ren, Yuxiu Lin, Xiao Yu, Hui Liu. Multi-Layer Contrastive Multi-View Clustering via Spectral Embedding Similarity[J]. Journal of Computer Research and Development, 2026 , 63 (4) : 960 -972 . DOI: 10.7544/issn1000-1239.202550353
近些年,随着信息技术的发展,数据采集的方式日益丰富。这些数据可能来源于不同角度或者不同传感器,被称为多视图数据。相较于传统的单视图数据,多视图数据中包含同一对象的多样化互补信息,可以更好地揭示数据中存在的潜在模式[1]。多视图聚类(multi-view clustering,MVC)旨在对多视图数据进行聚类,以挖掘视图间的一致性与互补性信息,并构建出能反映数据本质的聚类结构[2-3]。目前,多视图聚类已广泛应用于各种领域,如图像分割[4]、智能医疗[5]、行为识别[6]等。
对比学习(contrastive learning,CL)[7]是一种典型的无监督学习方法,其基本思想是通过最大化正样本间相似度和最小化负样本间相似度来学习特征表示。得益于正负样本对比优化机制,对比学习能有效增强模型的特征判别能力。在这个背景下,对比多视图聚类(contrastive multi-view clustering)作为对比学习的重要扩展,受到研究者们的广泛关注[8-10]。对比多视图聚类通常将同一样本的不同视图特征视为正样本,将其他样本特征视为负样本在跨视图层面进行特征学习。尽管取得了不错的聚类性能,但其在多视图聚类中的应用仍存在一些挑战:一方面,正样本数量不足使模型难以充分学习各簇的特征;另一方面,负样本的构造缺乏标签信息,存在假阴性样本的问题。
针对这些问题,Cui等人[11]提出了一种基于k近邻和伪标签的正负样本划分方法。简言之,对于每个样本,其k个最邻近的样本中伪标签一致的被视为正样本,其他样本则被视为负样本。Yang等人[12]提出用高置信度伪标签的方式构造正负样本。上述方法在一定程度上缓解了假阴性样本的问题并扩充了正样本对的数量,但同时也不可避免地产生了假阳性样本。Yan等人[13]在划分负样本时引入全局相似度关系矩阵,选择低相似度的样本来构造负样本。Yang等人[14]设计了一种具有动态梯度调节机制的对比学习模块。具体来说,当特征距离小于自适应阈值时,模型将反转梯度方向以拉近这些可能为假阴性的样本;当距离处于中间范围时,模型会抑制梯度幅值,防止过度优化。上述方法虽然缓解了负样本中假阴性的问题,但是忽略了对正样本对的构造。现有大部分对比多视图聚类方法都聚焦于解决其中一种问题,未能实现正负样本构造的整体优化,从而限制了聚类性能的提升。
为解决上述问题,本文提出基于谱嵌入相似度的多层对比多视图聚类(multi-layer contrastive multi-view clustering via spectral embedding similarity,MCSES)方法。与现有基于伪标签或邻居信息的对比多视图聚类方法不同,本文方法利用谱嵌入相似度矩阵提供的全局结构信息,在跨视图和全局视图层面进行对比学习,实现聚类性能提升。
具体来说,拉普拉斯矩阵的特征向量能有效反映数据的全局几何结构,并将数据映射到低维嵌入空间[15]。在该嵌入空间中,同一簇的样本高度聚集,不同簇间样本差异则更加显著,从而揭示出数据的潜在聚类结构。尽管谱嵌入方法能够取得较好的聚类效果,但其低维表示在细粒度特征区分方面仍存在局限性。相比之下,对比学习在优化正负样本时能够深入挖掘样本间细微的特征差异,但往往受到错误样本的干扰。本文利用谱嵌入相似度矩阵为模型提供全局先验信息,有效缓解了错误样本的干扰。同时,对比学习通过优化特征表示进一步提高谱嵌入相似度矩阵的质量,实现了两者优势的互补。此外,得益于谱嵌入相似度矩阵提供的全局先验信息,MCSES克服了传统对比学习仅能在跨视图层面构建正负样本对的限制。具体而言,在跨视图层面,MCSES进行对比学习来共享语义对齐;在全局视图层面中,MCSES通过可学习权重将多视图特征自适应融合后,再次进行对比学习,以强调全局一致性信息。MCSES网络结构如图1所示。
本文主要贡献包括3个方面:
1)提出基于谱嵌入相似度的对比学习策略。利用拉普拉斯矩阵的特征向量所反映的全局结构信息,缓解了因正负样本划分不当而导致的假阴性、假阳性问题。
2)设计了MCSES,可以充分探索每个视图共识信息并强调全局一致性信息,为下游聚类任务提供更具判别力的特征表示。
3)在6个公开数据集上进行了大量实验。结果表明,与其他先进多视图聚类方法相比,MCSES具有更优越的聚类性能。
本文将多视图聚类方法分为4类:1)基于子空间的多视图聚类[16]。此类方法侧重于在多视图间学习一致的子空间表示。Cai等人[17]提出在自表示矩阵中分解多视图的共性与差异,并结合多视图局部结构的联合平滑与低秩约束来获得更鲁棒的聚类表示。Ji等人[18]提出通过锚点表示取代全局自表示以提升大规模多视图场景下的聚类效率。2)基于矩阵分解的多视图聚类[19]。非负矩阵分解将每个视图分解为低秩矩阵,然后在低维空间中对数据进行聚类。Wei等人[20]提出提一种基于深度矩阵分解的多视图聚类方法,其中多视图数据矩阵被逐层分解为多个表示子空间。3)基于图的多视图聚类[21-23]。该方法为每个视图构建特定的图结构,然后通过正则化约束获得共享图,最后在共享图上利用谱聚类或其他算法获得聚类结果。例如,Fan等人[24]利用图自动编码器学习聚类特征,先选择最具信息量的图视图进行单一编码,再通过多解码器重构所有图视图结构。Tang等人[25]提出了一种基于交叉视图图扩散的多视图聚类方法,通过在不同视图间迭代扩散相似度,获得统一的聚类图。这3类多视图聚类方法也被统称为传统方法,其计算复杂度高,且难以有效捕捉到高维数据中存在的复杂特征表示,因此在实际应用场景中表现受限。4)深度多视图聚类。随着深度学习的发展,基于深度神经网络的多视图聚类方法逐渐成为主流。深度学习算法具有很强的非线性拟合能力,能够使用深度结构从大规模数据中进行深层特征学习,还能表达更加复杂的目标函数,进而提升聚类任务的性能。Xu等人[26]通过对多视图互信息的调控实现了视图公共特征与视图特有特征的解耦。Dong等人[27]提出跨视图拓扑图机制来协调多视图一致性与互补性信息。
近年来,为进一步挖掘多视图数据的潜在关联,对比学习被引入多视图聚类领域。例如,Wu等人[28]设计了一种自加权对比融合机制,让各视图的对比学习权重能够自适应调节。Luo等人[29]提出一种数据层融合的对比多视图聚类方法,避免了为各视图单独构建深度网络,简化了模型结构。Zhang等人[30]提出了一种基于最优传输的对比多视图聚类方法,通过在共同空间进行全局语义对齐并引入语义感知加权的对比学习机制,有效提升多视图聚类性能。尽管对比多视图聚类方法在性能上取得了显著提升,但正负样本划分不准确引发的假阴性、正样本不足等问题,仍在一定程度上限制了性能提升。
为此,本文提出MCSES方法,根据谱嵌入相似度衡量正负样本关系,利用低维嵌入空间中全局信息缓解样本划分错误的问题。
在本文中,多视图数据表示为$ X = \{ {{\boldsymbol X^v} \in {\mathbb{R}^{N \times {D_v}}}} \}_{v = 1}^V $,其中$V$表示视图总数,$ N $表示样本数,${D_v}$表示第$ v $个视图中输入数据的特征维数。为了方便描述,表1中列出本文常用的符号及定义。
谱嵌入方法通过对图拉普拉斯矩阵进行特征分解,将样本从原始空间映射到低维嵌入空间。具体而言,给定一个样本集合,通过构造图并计算其相似度矩阵${\boldsymbol {G}}$与度矩阵${\boldsymbol {D}}$,可以得到图的拉普拉斯矩阵:${\boldsymbol {L}} = {\boldsymbol {D}} - {\boldsymbol {G}}$。然后通过求解以下优化问题来获得低维嵌入矩阵${\boldsymbol {U}}$:
$ \mathop {\min }\limits_{\boldsymbol {U}} {\rm{tr(}}{{\boldsymbol {U}}^{\rm T}}{\boldsymbol {LU}}{{)}}\;\; {\text{s}}{\text{.t}}{\text{. }} {{\boldsymbol {U}}^{\rm T}}{\boldsymbol {U}} = {\boldsymbol {I}} \; ,$
其中${\boldsymbol {U}}$为嵌入矩阵。式(1)可等价表示为
$ \mathop {\min }\limits_{\boldsymbol {U}} \displaystyle\sum_{i,j}^{} {{G_{ij}}} { \| {{{\boldsymbol {z}}_i} - {{\boldsymbol {z}}_j}} \|^2} \; ,$
其中${{\boldsymbol {z}}_i}$为第$ i $个样本在谱嵌入空间中的特征表示,${G_{ij}}$表示样本$ i $与样本$j$之间的相似度。直观来看,式(2)鼓励图中相似度高的样本在嵌入空间中尽可能接近,相似度低的样本自然分离。这种机制实现了类内样本聚集与类间样本分离,从而隐式地揭示了数据的潜在聚类结构。基于此,谱嵌入相似度$ {\boldsymbol {M}} = {\boldsymbol {U}}{{\boldsymbol {U}}^{\rm T}} $可以有效捕获到全局簇结构,进而可以指导正负样本对的划分。
对于多视图聚类任务,模型从不同维度的多视图数据中提取出一致性与互补性的特征信息是至关重要的。许多研究已证明自编码器在特征提取方面的优异性能[31-32]。基于此,本文设计了各视图特定的自编码器模块,其中编码器提取各视图的深度特征,解码器则确保这些特征保留原始数据信息的同时减少噪声和冗余。设${f^v}$与$ {g^v} $分别表示第$ v $个视图的编码器和解码器。第$ v $个视图的编码器提取第$ i $个样本的潜在特征可以表示为
$ {\boldsymbol {z}}_i^v = {f^v}({\boldsymbol {x}}_i^v,{\theta ^v}) \; ,$
其中$ {\boldsymbol {z}}_i^v \in {\mathbb{R}^{{d_v}}} $表示第$ v $个视图第$ i $个样本的特征,${d_v}$表示第$ v $个视图的特征维度,${\theta ^v}$是编码器网络参数。同理,解码过程可表示为
$ {\boldsymbol {\hat x}}_i^v = {g^v}({\boldsymbol {z}}_i^v,{\varphi ^v}) \; ,$
其中$ {\boldsymbol {\hat x}}_i^v $表示第$ v $个视图第$ i $个样本的重构数据,${\varphi ^v}$是解码器网络参数。模型的重构损失计算公式为:
$ {\mathcal{L}_{\text{r}}} = \displaystyle\sum_{v = 1}^V {\displaystyle\sum_{i = 1}^N { \| {{\boldsymbol {x}}_i^v - {\boldsymbol {\hat x}}_i^v} \|_2^2} } \; 。$
多视图数据中,不同视图所包含信息的重要程度存在差异。为了充分挖掘视图间互补信息,本文设计了一种自适应加权融合模块,对自编码器输出的各视图特征进行自适应加权融合。该模块核心思想是利用可学习的权重参数,为每个视图分配不同的系数,从而凸显关键信息并抑制冗余噪声。加权融合的过程形式化为
$ {{\boldsymbol {z}}_i} = \displaystyle\sum_{v = 1}^V {{w_v}{\boldsymbol {z}}_i^v = \displaystyle\sum_{v = 1}^V {\dfrac{{{\exp(}{a_v}{)}}}{{\displaystyle\sum_{l = 1}^V {\exp ({a_l})} }}{\boldsymbol {z}}_i^v} } \; ,$
其中$ {{\boldsymbol {z}}_i} $表示第$ i $个样本的融合特征,${w_v}$表示第$ v $个视图的权重,由可学习的参数${a_v}$通过Softmax函数计算得到,各权重满足${w_v} \gt 0$且$ \displaystyle\sum_{v = 1}^V {{w_v} = 1} $。这些权重在训练过程中通过反向传播自动优化,能动态反映不同视图的重要性。
MCSES提取局部与全局特征后,引入基于谱嵌入相似度的多层对比学习机制进一步优化样本特征。具体来说,通过谱嵌入相似度矩阵引入全局先验信息以减少样本划分偏差,并借助多层对比学习同时建模跨视图与全局视图特征,全面提升聚类性能。
在基于对比学习的多视图聚类任务中,正负样本对的精确构造是实现高效特征学习的关键。已有研究表明,引入邻居信息有助于提高多视图聚类性能[33]。因此,现有对比多视图聚类方法往往引入k近邻或高阶随机游走的方法划分正负样本[34],但这些方法主要依赖局部信息,忽略了数据中各簇内部的全局结构信息。受谱聚类与文献[11]的启发,本文提出一种基于谱嵌入相似度的正负样本划分模块,其核心思想在于利用拉普拉斯矩阵特征向量能够反映数据全局结构这一特性,为正负样本划分提供精准的全局先验信息。具体来说,MCSES根据特征空间中各样本间的欧氏距离构造拉普拉斯矩阵。以第$ v $个视图为例,拉普拉斯矩阵的计算过程为:
$ {\boldsymbol {D}}_{}^v = {\rm diag}\left(\displaystyle\sum_{j = 1}^N {G_{ij}^v} \right) \; ,$
$ {{\boldsymbol {L}}^v} = {{\boldsymbol {D}}^v} - {{\boldsymbol {G}}^v} \; ,$
其中$ {{\boldsymbol {G}}^v} \in {\mathbb{R}^{N \times N}} $为第$ v $个视图样本间的相似度矩阵,$ {{\boldsymbol {D}}^v} \in {\mathbb{R}^{N \times N}} $为${{\boldsymbol {G}}^v}$的度矩阵。然后,MCSES对式(8)中的拉普拉斯矩阵进行谱分解,选取前$k$(聚类数)个最小的非零特征值对应的特征向量构成谱嵌入空间的特征矩阵,最后利用该特征矩阵构造谱嵌入相似度矩阵,即
$ {{\boldsymbol {L}}^v} = {{\boldsymbol {U}}^v}{{\boldsymbol {\varLambda }}^v}{({{\boldsymbol {U}}^v})^{\rm T}} \; ,$
$ {{\boldsymbol {M}}^v} = {\boldsymbol {U}}_k^v{({\boldsymbol {U}}_k^v)^{\rm T}} \; ,$
其中式(9)表示第$ v $个视图的拉普拉斯矩阵的谱分解。${{\boldsymbol {\varLambda }}^v} \in {\mathbb{R}^{N \times N}}$为由其特征值构成的对角矩阵。$ {\boldsymbol {U}}_k^v \in {\mathbb{R}^{N \times k}} $为前$k$个最小的非零特征值对应特征向量构成的特征矩阵。$ {{\boldsymbol {M}}^v} $为第$ v $个视图的谱嵌入相似度矩阵。本文引入阈值参数$f$,将相似度低于阈值$f$的样本对视为负样本对,将高置信度样本对视为正样本对。以全局视图特征为例,加权融合后第$ i $个样本对应的正负样本集合可以表示为
$ {P_i} = \{ j|{M_{ij}} \gt f\} \; ,$
$ {N_i} = \{ j|{M_{ij}} \leqslant f\} \; 。$
在跨视图正负样本划分中,MCSES首先将所有视图的特征拼接成统一的特征矩阵,然后构造跨视图谱嵌入相似度矩阵$ {{\boldsymbol {M}}^{\text{c}}} $。该过程与2.4.1节视图内正负样本对划分策略基本一致,但额外要求将同一样本在不同视图下提取的特征明确归为正样本对,以确保跨视图信息的一致性。因此,第$ i $个样本的跨视图正负样本集合可以表示为
$ P_i^{\text{c}} = \{ j|{M_{ij}} \gt f \vee j \in {\zeta _i}\} \; ,$
$ N_i^{\text{c}} = \{ j|{M_{ij}} \leqslant f \wedge j \notin {\zeta _i}\} \; ,$
其中${\zeta _i}$表示样本$ i $在所有视图中的索引的集合。
设对比学习损失矩阵为$ {\boldsymbol {K}} $,${K_{ij}}$表示为
$ {K_{ij}} = - \log \dfrac{{\exp \left(\dfrac{{S({{\boldsymbol {z}}_i},{{\boldsymbol {z}}_j})}}{\tau }\right)}}{{\displaystyle\sum_{n = 1}^N {\exp \left(\dfrac{{S({{\boldsymbol {z}}_i},{{\boldsymbol {z}}_n})}}{\tau }\right)} }} \; ,$
其中$\tau $为温度参数,$S(\;)$表示相似性函数。$ S({{\boldsymbol {z}}_i},{{\boldsymbol {z}}_j}) = \dfrac{{{\boldsymbol {z}}_i^{\rm T}{{\boldsymbol {z}}_j}}}{{ \| {{{\boldsymbol {z}}_i}} \| \| {{{\boldsymbol {z}}_j}} \|}} $。
跨视图对比损失可以表示为
$ \mathcal{L}_{\text{inter}}=\dfrac{1}{VN}\left(\displaystyle\sum_i^{VN}\dfrac{\displaystyle\sum_{j\in P_i^{\text{c}}}^{ }M_{ij}^{\text{c}}K_{ij}}{|P_i^{\text{c}}|}-\displaystyle\sum_i^{VN}\dfrac{\displaystyle\sum_{j\in N_i^{\text{c}}}^{ }K_{ij}}{|N_i^{\text{c}}|}\right)\; , $
其中$|P_i^{\text{c}}|$,$|N_i^{\text{c}}|$分别表示第$ i $个样本对应的正负样本对数量。$M_{ij}^{\text{c}}$表示第$ i $个样本与其正样本$j$之间的相似度。$V$为视图数。$N$为样本数。同理,加权融合后的全局视图下对比损失可以表示为
$ {\mathcal{L}_{{\text{intra}}}} = \dfrac{1}{N}\left(\displaystyle\sum_i^N {\dfrac{{\displaystyle\sum_{j \in {P_i}} {{M_{ij}}{K_{ij}}} }}{{|{P_i}|}}} - \displaystyle\sum_i^N {\dfrac{{\displaystyle\sum_{j \in {N_i}} {{K_{ij}}} }}{{|{N_i}|}}} \right)\; 。$
多层对比学习的损失函数可以表示为
$ {\mathcal{L}_{{\text{con}}}} = \beta {\mathcal{L}_{{\text{inter}}}} + (1 - \beta ){\mathcal{L}_{{\text{intra}}}} \; ,$
其中$\beta $是超参数,是平衡视图间对比学习损失$ {\mathcal{L}_{{\text{inter}}}} $与全局对比学习损失$ {\mathcal{L}_{{\text{intra}}}} $的权重。
为了加速模型收敛,本文采用“预训练+微调”的2阶段训练策略优化网络。
在预训练阶段,模型主要关注各视图编码器与解码器来获得较好的初始化参数${\theta ^v}$,${\varphi ^v}$。因此,预训练阶段的目标是最小化式(5)的重构损失。
在微调阶段,模型的目标是最小化式(5)中的重构损失与式(18)中的多层对比学习损失,来使模型在保持原始数据结构的同时,更好地捕获不同视图之间以及全局特征间的一致性与互补性信息。具体来说,微调阶段,模型通过最小化式(19)来更新网络参数。
$ {\mathcal{L}_{{\text{ft}}}} = \alpha {\mathcal{L}_{\text{r}}} + \beta {\mathcal{L}_{{\text{inter}}}} + (1 - \beta ){\mathcal{L}_{{\text{intra}}}} \; ,$
其中$ \alpha $,$\beta $是超参数,是用来平衡重构损失与对比学习损失的权重。具体的参数设置与实验结果将在第3节中详细讨论。算法1提供了本文模型的训练流程。
算法1. 训练算法。
输入:多视图数据$X$,聚类数量$K$,特征维度$d$,超参数$ \alpha $,$\beta $,温度系数$\tau $;
输出:聚类结果。
① 通过最小化式(5)初始化自编码器参数;
② while 未收敛
③ 通过自编码器提取各视图特征$ {\boldsymbol {z}}_{}^v $;
④ 通过式(6)计算全局特征$ {\boldsymbol {z}} $;
⑤ 通过式(11)~(14)计算跨视图与全局正负样 本对;
⑥ 通过式(5)计算重构损失${\mathcal{L}_{\text{r}}}$;
⑦ 通过式(18)计算多层对比学习损失${\mathcal{L}_{{\text{con}}}}$;
⑧ 通过式(19)计算模型整体损失${\mathcal{L}_{{\text{ft}}}}$;
⑨ 对整个网络进行反向传播与训练;
⑩ end while
⑪ 得到最终特征矩阵${\boldsymbol {Z}}$,执行k-means聚类,返回聚类结果。
1)实验环境
本文提出的模型MCSES在Windows11操作系统上实现,采用Python3.9.21与Pytorch2.2.2进行部署。主要硬件参数为NVIDIA GeForce RTX 4060 GPU,8 GB显存,Intel® CoreTM i7-13650HX,24 GB内存。
2)实验参数
在所有数据集上,各视图编码器统一设置为(500,1000,2000,128),解码器统一设置为(128,2000,1000,500)。ReLU函数用作激活函数。Adam优化器对网络进行优化。学习率设置为0.0003,批次大小为256。在预训练阶段,训练迭代次数设置为200。温度系数$\tau $在Cora,Digit-Product,RGB-D数据集上统一设置为0.5,在synthetic3d数据集上设置为0.1,在LandUse-21上设置为0.3,在MNIST-USPS数据集上设置为1。平衡重构损失权重的超参数$\alpha \in${0.0001,0.001,0.01,0.1,1}。平衡跨视图与全局对比学习损失权重的超参数$\beta \in${0.1,0.2,0.5,0.6,0.8,0.9,1},阈值参数$f\in${0.2,0.5,0.6,0.9}。
3)评价指标
本文使用4种通用的评价指标综合评价方法的聚类性能,包括准确性(accuracy,ACC)、标准化互信息(normalized mutual information,NMI)、调整兰德指数(adjusted Rand index,ARI)、纯度(purity,PUR)。对于每一个评价指标,其数值越大意味着聚类结果越好。
4)数据集
实验在表2所示的6个公开数据集上进行。其中Cora[35]是一个广泛使用的学术文献引用数据集,包含2708个文档和7个类别。每篇文档包含内容、入站、出站与引用4个视图。LandUse-21[36]数据集包含21个类别的2100幅遥感图像,采用PHOG和LBP特征作为2个视图。MNIST-USPS[37],Digit-Product[26]均为书写数字数据集,分别涵盖2种不同风格的数字图像。RGB-D[38]是一个室内场景图像数据集,包含5000个样本与13个类别。Synthetic3d[39]数据集为一个合成多视图数据集,包含600个样本和3个类别,每个样本在3个视图下具有不同的特征表示。
5)对比方法
为了全面评估MCSES的效果,将之与9种先进的多视图聚类方法进行比较,这些方法可分为3类:深度多视图聚类(DSMVC[40],DFMVC[41])、深度对比多视图聚类(SCMVC[28],SCM[29],MFLVC[42])、优化了正负样本划分策略的深度对比多视图聚类(DCMVC[11],DealMVC[12],GCFAgg[13],CSOT[30]),所有对比方法都遵循原文建议进行参数调整。
实验结果如表3和表4所示。从表中实验结果可以看出,MCSES在所有数据集上均取得良好的聚类性能,尤其在Cora与RGB-D数据集的ACC指标上,分别比次优方法提高了约4个百分点和3.9个百分点。与深度多视图聚类方法DSMVC相比,MCSES在各数据集上均展现出优势。与3种对比多视图聚类方法SCM,MFLVC,SCMVC相比,MCSES取得了更好的效果。因为传统对比多视图聚类方法在负样本构造上存在假阴性的问题,导致同一类样本在特征空间被错误地拉开距离。MCSES基于谱嵌入相似度对正负样本进行划分,有效缓解了这个问题,并且优于最新的优化了正负样本划分策略的深度对比多视图聚类方法DCMVC,DealMVC,GCFAgg,CSOT。这证实了基于谱嵌入相似度的对比学习方法的有效性。此外,MCSES采用多层次的对比学习策略,对不同视图的潜在特征进行逐层建模,不仅能够充分挖掘各视图所包含的差异化信息,还能在全局表征层面强化跨视图之间的一致性,实现聚类性能的提升。
为分析MCSES的计算复杂度,设每个数据集的视图数量为$V$,每个视图的特征维度为$D$,聚类类别数为$K$,训练批量大小为$B$。每次训练迭代主要包括重构阶段与对比学习阶段。在重构阶段,模型对每个视图分别进行编码与解码,计算复杂度为$O(V BD)$。对比学习阶段主要包括特征相似度构建与谱嵌入掩码生成2部分。其中,相似度构建的计算复杂度为$O(V{B^2}D)$;谱嵌入掩码包括拉普拉斯矩阵构建及其特征分解,其复杂度为$O(V{B^3})$,以及谱相似度矩阵生成$ O(V{B^2}K) $。基于此,整体复杂度可合并表达为$O(V({B^3} + {B^2}D + {B^2}K))$。
虽然MCSES在多视图聚类任务中表现出良好的建模能力,但谱分解过程带来的时间复杂度为$O({B^3})$,限制了MCSES在大规模数据集上的训练效率。为缓解上述问题,未来研究考虑引入近似谱方法,在保证嵌入效果的同时降低谱分解计算开销,以提升模型在大规模数据集的训练效率。
为了直观研究MCSES的有效性,图2中展示了Digit-Product与MNIST-USPS数据集上通过t-SNE算法降维得到的特征结果。显然,随着迭代次数的增加,聚类内部数据变得紧凑,不同聚类间距离逐渐拉大,聚类结构变得清晰。这表明,本文所设计的基于谱嵌入相似度的多层对比学习机制,有效缓解了错误样本的干扰,提升了潜在空间中簇内的紧凑性与簇间的分离性,为聚类任务提供更具判别力的特征表示。
本文方法包含$ \alpha $,$\beta $,$ \tau $,$f$等4个超参数,以LandUse-21和Cora数据集为例,本节采取网格搜索策略来选择最佳的参数组合。图3展示了不同$ \alpha $,$\beta $取值下得到的聚类结果。可以发现,MCSES在LandUse-21数据集上的聚类性能在超参数$ \alpha $,$\beta $变化下表现出一定的波动。在$ \alpha\in[0.1\;\;1] $,$\beta \in[0.1\;\;0.2]$的范围内,MCSES能够取得较优的ACC和NMI。LandUse-21数据集上超参数分别被设置为$ \alpha $=1和$\beta $=0.1。相比之下,Cora数据集上的聚类结果相对稳定。超参数分别被设置为$ \alpha $=0.0001和$\beta $=0.8。
图4展示了不同$\tau $,$f$取值下得到的聚类结果。阈值参数$f$控制着正负样本的划分。$f$过小会引入噪声,导致聚类性能不佳;$f$过大会使基于谱嵌入相似度的对比多视图聚类退化为传统的对比多视图聚类。
依据实验数据,LandUse-21与Cora数据集上的$f$分别被设置为0.6和0.5。此外,当温度系数$\tau $从0.1变化到1时,MCSES聚类性能呈现先升后降的趋势。根据经验,温度系数$\tau $在Cora,Digit-Product,RGB-D数据集上统一设置为0.5,在Synthetic3d数据集上设置为0.1,在LandUse-21上设置为0.3,在MNIST-USPS数据集上设置为1。
为进一步探究各模块对模型的贡献,本节设计了3组消融实验。表5展示了不同损失项对Cora与LandUse-21数据集聚类性能的影响。结果显示,仅使用重构损失(方案A)时,模型在2个数据集上各项指标均较低。在Cora数据集上引入跨视图对比损失$ {\mathcal{L}_{{\text{inter}}}} $(方案C)后,模型各项指标显著提高,而在LandUse-21数据集上性能提升有限。在LandUse-21数据集上引入全局对比损失$ {\mathcal{L}_{{\text{intra}}}} $(方案D)后模型性能显著提高,而在Cora数据集上模型性能甚至低于仅使用重构损失的情况。这是由于在Cora数据集上,超参数$\beta $设为0.8使得跨视图对比损失主导模型优化,而在LandUse-21数据集上,$\beta $值设为0.1,模型主要侧重于全局特征的学习。当同时引入$ {\mathcal{L}_{\text{r}}} $,$ {\mathcal{L}_{{\text{inter}}}} $,$ {\mathcal{L}_{{\text{intra}}}} $(方案E)后,模型在2个数据集上均取得最佳性能。方案B则将基于谱嵌入相似度的对比学习替换为传统对比学习,实验结果显示其性能不如方案C,进一步证明了谱嵌入相似度矩阵的作用。
图5展示了自适应加权融合模块对聚类性能的提升效果。实验结果表明,引入该模块后,模型的各项聚类指标均优于采用简单平均的模型。这些结果证明了自适应加权融合模块在提升聚类性能方面的有效性。总体来看,删除任一模块均会导致聚类性能下降,充分证明MCSES中各模块均对聚类性能的提升有积极作用。
在现实应用场景中,多视图数据采集过程常常会出现部分视图缺失的情况,这将直接影响模型对样本语义信息的完整性建模。为评估MCSES在现实应用场景下的鲁棒性,本文设计了缺失率为10%,20%,30%下的模拟实验,随机剔除部分样本中的某一个视图,实验结果如表6所示。
结果表明,MCSES在各数据集上仍能保持较高的聚类性能,表现出对部分视图缺失具有一定的容错能力。但随着缺失率进一步升高,MCSES性能逐渐下降,说明模型在高缺失率的场景下鲁棒性不足。具体而言,一方面,视图缺失显著减少了可用于构造跨视图正样本对数量,削弱了模型获取判别性特征的能力;另一方面,缺失视图破坏了多视图间的结构一致性,导致谱嵌入的相似性关系失真,影响正负样本划分的准确性。
总体而言,MCSES在轻度视图缺失下表现出一定的鲁棒性,但面对高缺失场景下会出现性能退化的问题。未来考虑引入视图补全机制,以增强模型在复杂场景下的适应能力。
图6展示了MCSES在MNIST-USPS数据集上的收敛曲线。从图中可以看出,随着迭代次数的增加,损失值持续下降直至稳定收敛;而ACC和NMI指标值逐步上升并最终趋于稳定。这些结果证明了MCSES方法的良好收敛性。
本文提出了一种基于谱嵌入相似度的多层对比多视图聚类方法(MCSES)。与现有的对比多视图聚类方法不同,基于谱嵌入相似度的正负样本划分机制使模型可以充分学习各簇全局特征,同时缓解了现有划分方法存在的错误样本问题。此外,多层对比学习框架使MCSES可以充分探索每个视图共识信息,并进一步学习全局特征表示。在多个公开数据集上进行了大量实验证明了MCSES具有更优越的聚类性能。然而,在现实场景中,数据可能会出现视图缺失、部分类别缺失、样本数较少或数据分布不均等监督信息不足问题,这将直接影响谱嵌入构建的稳定性,进而降低正负样本对划分的可靠性。为此,后续将深入研究不完整多视图聚类问题,探索在视图缺失场景下如何构建鲁棒的谱嵌入引导机制,从而提升模型在现实场景中的泛化能力。
作者贡献声明:任晓阳负责模型的设计和实现以及论文的撰写和修改;林毓秀、于晓负责实验的整理与分析;刘慧提出研究方向,把握论文的创新性,并指导论文修改。

参考文献 引证文献
排序方式:
1
刘新旺. 多视图学习的基础理论和方法研究[J]. 中国基础科学, 2022, 24(3): 27−34
Liu Xinwang. Research on the fundamental theory and methods of multi-view learning[J]. China Basic Science, 2022, 24(3): 27−34 (in Chinese)
2
赵玉涵, 陈松灿. 消除互补性争议的多视图聚类算法[J]. 计算机研究与发展, 2025, 62(5): 1216−1225
Zhao Yuhan, Chen Songcan. Multi-view clustering algorithm via eliminating the complementarity controversy[J]. Journal of Computer Research and Development, 2025, 62(5): 1216−1225 (in Chinese)
3
Chao Guoqing, Sun Shiliang, Bi Jinbo. A survey on multiview clustering[J]. IEEE Transactions on Artificial Intelligence, 2021, 2(2): 146−168
4
张荣国, 曹俊辉, 胡静, 等. 基于非负正交矩阵分解的多视图聚类图像分割算法[J]. 模式识别与人工智能, 2023, 36(6): 556−571
Zhang Rongguo, Cao Junhui, Hu Jing, et al. Non-negative orthogonal matrix factorization based multi-view clustering image segmentation algorithm[J]. Pattern Recognition and Artificial Intelligence, 2023, 36(6): 556−571(in Chinese)
5
Zhang Yilin, Chen Shuo, Mccoy R G, et al. Multi-view k-nearest neighbor graph contrastive learning on multi-modal biomedical data[C]//Proc of the 22nd Int Conf on Artificial Intelligence in Medicine. Berlin: Springer, 2024: 238−248
6
Yan Xiaoqiang, Gan Yingtao, Mao Yiqiao, et al. Live and learn: Continual action clustering with incremental views[C]//Proc of the 38th AAAI Conf on Artificial Intelligence. Palo Alto, CA: AAAI, 2024: 16264−16271
7
Wu Zhirong, Xiong Yuanjun, Yu S X, et al. Unsupervised feature learning via non-parametric instance discrimination[C]//Proc of the IEEE Conf on Computer Vision and Pattern Recognition. Piscataway, NJ: IEEE, 2018: 3733−3742
8
Bian Jintang, Xie Xiaohua, Lai Jianhuang, et al. Multi-view contrastive clustering via integrating graph aggregation and confidence enhancement[J]. Information Fusion, 2024, 108: 102393
9
Wang Haiyue, Wang Quan, Miao Qiguang, et al. Joint learning of data recovering and graph contrastive denoising for incomplete multi-view clustering[J]. Information Fusion, 2024, 104: 102155
10
Liu Jiyuan, Liu Xinwang, Yang Yuexiang, et al. Contrastive multi-view kernel learning[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023, 45(8): 9552−9566
11
Cui Jinrong, Li Yuting, Huang Han, et al. Dual contrast-driven deep multi-view clustering[J]. IEEE Transactions on Image Processing, 2024, 33: 4753−4764
12
Yang Xihong, Jin Jiaqi, Wang Siwei, et al. DealMVC: Dual contrastive calibration for multi-view clustering[C]//Proc of the 31st ACM Int Conf on Multimedia. New York: ACM, 2023: 337−346
13
Yan Weiqing, Zhang Yuanyang, Lv Chenlei, et al. GCFAgg: Global and cross-view feature aggregation for multi-view clustering[C]//Proc of the IEEE/CVF Conf on Computer Vision and Pattern Recognition. Piscataway, NJ: IEEE, 2023: 19863−19872
14
Yang Mouxing, Li Yunfan, Hu Peng, et al. Robust multi-view clustering with incomplete information[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 45(1): 1055−1069
15
Belkin M, Niyogi P. Laplacian eigenmaps and spectral techniques for embedding and clustering[C]//Proc of the 15th Int Conf on Neural Information Processing Systems: Natural and Synthetic. Cambridge, MA: MIT, 2001: 585−591
16
林毓秀, 刘慧, 于晓, 等. 面向子空间聚类的多视图统一表示学习网络[J]. 计算机研究与发展, 2025, 62(5): 1248−1261
Lin Yuxiu, Liu Hui, Yu Xiao, et al. A multi-view unified representation learning network for subspace clustering[J]. Journal of Computer Research and Development, 2025, 62(5): 1248−1261(in Chinese)
17
Cai Xiaosha, Huang Dong, Zhang Guangyu, et al. Seeking commonness and inconsistencies: A jointly smoothed approach to multi-view subspace clustering[J]. Information Fusion, 2023, 91: 364−375
18
Ji Jintian, Feng Songhe. Anchor structure regularization induced multi-view subspace clustering via enhanced tensor rank minimization[C]//Proc of the IEEE Int Conf on Computer Vision. Piscataway, NJ: IEEE, 2023: 19343−19352
19
Liu Jialu, Wang Chi, Gao Jing, et al. Multi-view clustering via joint nonnegative matrix factorization[C]//Proc of the 2013 SIAM Int Conf on Data Mining. Philadelphia, PA: SIAM, 2013: 252−260
20
Wei Shaowei, Wang Jun, Yu Guoxian, et al. Multi-view multiple clusterings using deep matrix factorization[C]//Proc of the 34th AAAI Conf on Artificial Intelligence. Palo Alto, CA: AAAI, 2020: 6348−6355
21
Jing Peiguang, Su Yuting, Li Zhengnan, et al. Learning robust affinity graph representation for multi-view clustering[J]. Information Sciences, 2021, 544: 155−167
22
Pan Erlin, Kang Zhao. Multi-view contrastive graph clustering[C]//Proc of the 35th Int Conf on Neural Information Processing Systems. Cambridge, MA: MIT, 2021: 2148−2159
23
Li Lusi, He Haibo. Bipartite graph based multi-view clustering[J]. IEEE Transactions on Knowledge and Data Engineering, 2020, 34(7): 3111−3125
24
Fan Shaohua, Wang Xiao, Shi Chuan, et al. One2multi graph autoencoder for multi-view graph clustering[C]//Proc of the Web Conf 2020. New York: ACM, 2020: 3070−3076
25
Tang Chang, Liu Xinwang, Zhu Xinzhong, et al. CGD: Multi-view clustering via cross-view graph diffusion[C]//Proc of the 34th AAAI Conf on Artificial Intelligence. Palo Alto, CA: AAAI, 2020: 5924−5931
26
Xu Jie, Ren Yazhou, Tang Huayi, et al. Multi-VAE: Learning disentangled view-common and view-peculiar visual representations for multi-view clustering[C]//Proc of the IEEE Int Conf on Computer Vision. Piscataway, NJ: IEEE, 2021: 9234−9243
27
Dong Zhibin, Wang Siwei, Jin Jiaqi, et al. Cross-view topology based consistent and complementary information for deep multi-view clustering[C]//Proc of the IEEE Int Conf on Computer Vision. Piscataway, NJ: IEEE, 2023: 19440−19451
28
Wu Song, Zheng Yan, Ren Yazhou, et al. Self-weighted contrastive fusion for deep multi-view clustering[J]. IEEE Transactions on Multimedia, 2024, 26: 9150−9162
29
Luo Caixuan, Xu Jie, Ren Yazhou, et al. Simple contrastive multi-view clustering with data-level fusion[C]//Proc of the 33rd Int Joint Conf on Artificial Intelligence. San Francisco, CA: Morgan Kaufmann, 2024: 4697−4705
30
Zhang Qian, Zhang Lin, Song Ran, et al. Learning common semantics via optimal transport for contrastive multi-view clustering[J]. IEEE Transactions on Image Processing, 2024, 33: 4501−4515
31
Hinton G E, Salakhutdinov R R. Reducing the dimensionality of data with neural networks[J]. Science, 2006, 313(5786): 504−507
32
Coates A, Ng A, Lee H. An analysis of single-layer networks in unsupervised feature learning[C]//Proc of the 14th Int Conf on Artificial Intelligence and Statistics. Brookline, MA: JMLR W&CP, 2011: 215−223
33
Wen Jie, Zhang Zheng, Zhang Zhao, et al. Dimc-net: Deep incomplete multi-view clustering network[C]//Proc of the 28th ACM Int Conf on Multimedia. New York: ACM, 2020: 3753−3761
34
Lu Yiding, Lin Yijie, Yang Mouxing, et al. Decoupled contrastive multi-view clustering with high-order random walks[C]//Proc of the 38th AAAI Conf on Artificial Intelligence. Palo Alto, CA: AAAI, 2024: 14193−14201
35
Fang Siguo Huang Dong, Cai Xiaosha, et al. Efficient multi-view clustering via unified and discrete bipartite graph learning[J]. IEEE Transactions on Neural Networks and Learning Systems, 2023, 35(8): 11436−11447
36
Lin Yijie, Gou Yuanbiao, Liu Xiaotian, et al. Dual contrastive prediction for incomplete multi-view representation learning[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 45(4): 4447−4461
37
Peng Xi, Huang Zhenyu, Lv Jiancheng, et al. COMIC: Multi-view clustering without parameter selection[C]//Proc of the 36th Int Conf on Machine Learning. New York: ACM, 2019: 5092−5101
38
Kong Chen, Lin Dahua, Bansal M, et al. What are you talking about? text-to-image coreference[C]//Proc of the IEEE Conf on Computer Vision and Pattern Recognition. Piscataway, NJ: IEEE, 2014: 3558−3565
39
Kumar A, Rai P, Daume H. Co-regularized multi-view spectral clustering[C]//Proc of the 25th Int Conf on Neural Information Processing Systems. Cambridge, MA: MIT, 2011: 1413−1421
40
Tang Huayi, Liu Yong. Deep safe multi-view clustering: Reducing the risk of clustering performance degradation caused by view increase[C]//Proc of the IEEE/CVF Conf on Computer Vision and Pattern Recognition. Piscataway, NJ: IEEE, 2022: 202−211
41
Ren Yazhou, Pu Jingyu, Cui Chenhang, et al. Dynamic weighted graph fusion for deep multi-view clustering[C]//Proc of the 33rd Int Joint Conf on Artificial Intelligence. San Francisco, CA: Morgan Kaufmann, 2024: 4842−4850
42
Xu Jie, Tang Huayi, Ren Yazhou, et al. Multi-level feature learning for contrastive multi-view clustering[C]//Proc of the IEEE/CVF Conf on Computer Vision and Pattern Recognition. Piscataway, NJ: IEEE, 2022: 16051−16060
2026年第63卷第4期
89
41
0
引用本文
BibTeX
文章信息
doi: 10.7544/issn1000-1239.202550353
  • 接收时间:2025-05-26
  • 首发时间:2026-11-17
  • 出版时间:2026-04-01
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2025-05-26
  • 修回日期:2025-09-24
  • 录用日期:2025-10-11
基金
作者信息
    山东财经大学计算机与人工智能学院 济南 250014
    山东省数字经济轻量智算与可视化重点实验室(山东财经大学) 济南 250014

通讯作者:

参考文献
分享链接
https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN/10.7544/issn1000-1239.202550353
分享至
全文二维码

扫描看全文

引用本文
BibTeX
本文的引用情况
2种不同金属材料的力学参数
科
Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)
属
Genus
种数
Number of
species
占总种数比例
Percentage of total
species (%)
鹅膏菌科Amanitaceae 2 11 5.26 鹅膏菌属 Amanita 10 4.78
小菇科 Mycenaceae 2 12 5.74 丝盖伞属 Inocybe 5 2.39
多孔菌科 Polyporaceae 8 14 6.70 蜡蘑属 Laccaria 5 2.39
红菇科 Russulaceae 3 23 11.00 小皮伞属 Marasmius 6 2.87
小菇属 Mycena 11 5.26
光柄菇属 Pluteus 5 2.39
红菇属 Russula 17 8.13
栓菌属 Trametes 5 2.39
关闭全屏