典型文献
Contrastive Consensus Graph Learning for Multi-View Clustering
文献摘要：
Dear Editor,
This letter proposes a contrastive consensus graph learning model for multi-view clustering. Graphs are usually built to outline the cor-relation between multi-model objects in clustering task, and multi-view graph clustering aims to learn a consensus graph that integrates the spatial property of each view. Nevertheless, most graph-based models merely consider the overall structure from all views but neglect the local spatial consistency between diverse views, resulting in the lack of global spatial consistency in the learned graph. To overcome this issue, a deep convolutional network is built to explore latent local spatial information from raw affinity graphs. Specifically, we employ a consensus graph constraint to preserve the global con-sistency between the learned graph and raw graphs. Furthermore, a contrastive reconstruction loss is introduced to achieve the sample-level approximation between reconstructed graphs and raw graphs, which facilitates the network to enhance the consensus graph learn-ing. Experiments on six classical datasets demonstrate that the pro-posed model outperforms other nine state-of-the-art algorithms.
文献关键词：
中图分类号：
作者姓名：
Shiping Wang;Xincan Lin;Zihan Fang;Shide Du;Guobao Xiao
作者机构：
College of Computer and Data Science,Fuzhou University,Fuzhou 350108,China;College of Computer and Control Engineering,Minjiang University,Fuzhou 350108,China
文献出处：
引用格式：
[1]Shiping Wang;Xincan Lin;Zihan Fang;Shide Du;Guobao Xiao-.Contrastive Consensus Graph Learning for Multi-View Clustering)[J].自动化学报（英文版）,2022(11):2027-2030
A类：
sistency
B类：
Contrastive,Consensus,Learning,Multi,View,Clustering,Dear,Editor,
This,letter,proposes,contrastive,consensus,learning,multi,clustering,Graphs,are,usually,built,outline,cor,relation,between,objects,task,aims,that,integrates,spatial,property,each,Nevertheless,most,models,merely,consider,overall,structure,from,views,but,neglect,local,consistency,diverse,resulting,lack,global,learned,To,overcome,this,issue,deep,convolutional,network,explore,latent,information,raw,affinity,graphs,Specifically,employ,constraint,preserve,Furthermore,reconstruction,loss,introduced,achieve,sample,level,approximation,reconstructed,which,facilitates,enhance,Experiments,six,classical,datasets,demonstrate,posed,outperforms,other,nine,state,art,algorithms
AB值：
0.583528
相似文献
机标中图分类号，由域田数据科技根据网络公开资料自动分析生成，仅供学习研究参考。
