Self-weighted Multiview Clustering with Multiple Graphs

By Feiping Nie, Jing Li, Xuelong Li. International Joint Conference on Artificial Intelligence, 2017

608 citations recorded by the source corpus.

In multiview learning, it is essential to assign a reasonable weight to each view according to the view importance. Thus, for multiview clustering task, a wise and elegant method should achieve clustering multiview data while learning the view weights. In this paper, we propose to explore a Laplacian rank constrained graph, which can be approximately as the centroid of the built graph for each view with different confidences. We start our work with a natural thought that the weights can be learned by introducing a hyperparameter. By analyzing the weakness of this way, we further propose a new multiview clustering method which is totally selfweighted. More importantly, once the target graph is obtained in our models, we can directly assign the cluster label to each data point and do not need any postprocessing such as K-means in standard spectral clustering. Evaluations on two synthetic datasets indicate the effectiveness of our methods. Compared with several representative graphbased multiview clustering approaches on four realworld datasets, the proposed methods achieve the better performances and our new clustering method is more practical to use.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Urban Studies

View Paper: https://www.semanticscholar.org/paper/14a3931d5aec9ee91fa6894dff3a61179b002b1c

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Bing Liu
2. Hao Wang
3. Jiancheng Lv
4. Yan Yang
5. Cheng Deng
6. Xi Peng
7. Xu Yang
8. Joey Tianyi Zhou

In Computer Vision and Pattern Recognition

1. Bing Liu
2. Cheng Deng
3. Yan Yang
4. Hao Wang
5. Xu Yang
6. Jiancheng Lv
7. Xi Peng
8. Joey Tianyi Zhou

In Urban Studies

1. Jiancheng Lv
2. Xi Peng
3. Bing Liu
4. Hao Wang
5. Yan Yang
6. Feiping Nie
7. Cheng Deng
8. Yuanbiao Gou

Citations in the corpus

• Multi-view Contrastive Graph Clustering (2021, unknown): “To directly cluster multiple graphs, self-weighted multi-view clustering (SwMC) [Nie et al., 2017] method learns a shared graph from multiple graphs by using a novel weighting strategy.”
• COMPLETER: Incomplete Multi-view Clustering via Contrastive Prediction (2021, methodology): “As one of the most important unsupervised multi-view methods, multiview clustering (MVC) aims to separate data points into different clusters in an unsupervised fashion [11,17,20,29,40, 54].”
• GMC: Graph-Based Multi-View Clustering (2020, methodology): “Among these multi-view clustering methods, one representative category of methods is the graph-based methods [5], [6], [7], [8], [9], [10], [11], [12].”
• COMIC: Multi-view Clustering Without Parameter Selection (2019, methodology): “…clustering (Zhang et al., 2017), deep canonical correlation analysis (DCCA) (Andrew et al., 2013), self-weighted multi-view clustering (SwMVC) (Nie et al., 2017), deep canonically correlated autoencoders (DCCA) (Wang et al., 2015), and binary multi-view clustering (BMVC) (Zhang et al., 2018).”
• Multi-level Feature Learning for Contrastive Multi-view Clustering (2021, methodology): “The third category of MVC methods is graph based MVC [28, 34], where graph structures are build to preserve the adjacency relationship among samples.”
• Saliency Detection via a Multiple Self-Weighted Graph-Based Manifold Ranking (2020, background): “Straightforwardly, without the regularization term( γ → 0 ) [50], the trivial solution will be obtained, where the weight of best view will be assigned to 1 and other weights will be 0s.”
• A Survey on Multi-View Clustering (2017, unknown): “[33] assumes a common indicator matrix and then solves the problem by minimizing the differences between the common indicator matrix and each similarity matrix.”