Localized Sparse Incomplete Multi-view Clustering

  title={Localized Sparse Incomplete Multi-view Clustering},
  author={Chengliang Liu and Zhihao Wu and Jie Wen and Yong Xu and Chao Huang},
—Incomplete multi-view clustering, which aims to solve the clustering problem on the incomplete multi-view data with partial view missing, has received more and more atten- tion in recent years. Although numerous methods have been developed, most of the methods either cannot flexibly handle the incomplete multi-view data with arbitrary missing views or do not consider the negative factor of information imbalance among views. Moreover, some methods do not fully explore the local structure of all… 

Figures and Tables from this paper



Doubly Aligned Incomplete Multi-view Clustering

This paper proposes a Doubly Aligned Incomplete Multi-view Clustering algorithm (DAIMC) based on weighted semi-nonnegative matrix factorization (semi-NMF), which has two unique advantages: solving the incomplete view problem by introducing a respective weight matrix for each view; and reducing the influence of view incompleteness on clustering by enforcing the basis matrices of individual views being aligned with the help of regression.

Multiple Incomplete Views Clustering via Weighted Nonnegative Matrix Factorization with L2, 1 Regularization

The proposed MIC (Multi-Incomplete-view Clustering), an algorithm based on weighted nonnegative matrix factorization with L2,1 regularization, works by learning the latent feature matrices for all the views and generating a consensus matrix so that the difference between each view and the consensus is minimized.

One-Pass Incomplete Multi-view Clustering

An One-Pass Incomplete Multi-view Clustering framework (OPIMC), different from the existing and sole online IMC method, OPIMC can directly get clustering results and effectively determine the termination of iteration process by introducing two global statistics.

Partial Multi-view Subspace Clustering

This work proposes a novel multi-view clustering method, called Partial Multi-view Subspace Clustering (PMSC), that seeks the latent space and performs data reconstruction simultaneously to learn the subspace representation, leading to a more comprehensive data description.

Unified Embedding Alignment with Missing Views Inferring for Incomplete Multi-View Clustering

A locality-preserved reconstruction term is introduced to infer the missing views such that all views can be naturally aligned and a consensus graph is adaptively learned and embedded via the reverse graph regularization to guarantee the common local structure of multiple views.

Consensus Graph Learning for Multi-View Clustering

A novel multi-view clustering method that is able to construct an essential similarity graph in a spectral embedding space instead of the original feature space is proposed and an efficient optimization algorithm is designed to solve the resultant optimization problem.

Partial Multi-View Clustering using Graph Regularized NMF

The proposed method, which is referred to as GPMVC (Graph Regularized Partial Multi-View Clustering), is compared against 7 baseline methods (including PVC) on 5 publicly available text and image datasets and outperforms all baselines.

DIMC-net: Deep Incomplete Multi-view Clustering Network

DIMC-net designs several view-specific encoders to extract the high-level information of multiple views and introduces a fusion graph based constraint to explore the local geometric information of data.

Efficient and Effective Regularized Incomplete Multi-View Clustering

This paper proposes an Efficient and Effective Incomplete Multi-view Clustering (EE-IMVC) algorithm, which proposes to impute each incomplete base matrix generated by incomplete views with a learned consensus clustering matrix to address issues of intensive computational and storage complexities, over-complicated optimization and limitedly improved clustering performance.

Online multi-view clustering with incomplete views

An online multi-view clustering algorithm, OMVC, which deals with large-scale incomplete views is proposed, and dynamic weight setting is introduced to give lower weights to the incoming missing instances in different views to reduce the computational time.