Partial Multi-View Clustering

By Shaoyuan Li, Yuan Jiang, Zhi-Hua Zhou. AAAI Conference on Artificial Intelligence, 2014

451 citations recorded by the source corpus.

Real data are often with multiple modalities or comingfrom multiple channels, while multi-view clusteringprovides a natural formulation for generating clustersfrom such data. Previous studies assumed that each exampleappears in all views, or at least there is one viewcontaining all examples. In real tasks, however, it is oftenthe case that every view suffers from the missing ofsome data and therefore results in many partial examples,i.e., examples with some views missing. In this paper,we present possibly the first study on partial multiviewclustering. Our proposed approach, PVC, worksby establishing a latent subspace where the instancescorresponding to the same example in different viewsare close to each other, and similar instances (belongingto different examples) in the same view should bewell grouped. Experiments on two-view data demonstratethe advantages of our proposed approach.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Signal Processing

View Paper: https://www.semanticscholar.org/paper/03174fdee0ce3ae01717df8808594d1d55b4fcbc

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

In Computer Vision and Pattern Recognition

In Signal Processing

Citations in the corpus

Incomplete Multiview Spectral Clustering With Adaptive Graph Learning (2020, methodology): “The values of blood test and magnetic resonance images can be viewed as two references for disease diagnosing [3].”
Late Fusion Incomplete Multi-view Clustering (2018, background): “It has been intensively studied during the last few decade [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14] and widely used in various applications, including object segmentation [15], [16], object pose estimation [17], image re-ranking [18], saliency detection [19], information ”
Multi-view clustering: A survey (2018, unknown): “In addition, Li et al.[159] presented a partial multi-view clustering algorithm (named PVC), which is specifically designed for two-view datasets.”
Multiple Incomplete Views Clustering via Weighted Nonnegative Matrix Factorization with L2, 1 Regularization (2015, methodology): “As far as we know, [26,32] are the only two works that do not require the completeness of any view.”
Learning With Feature Evolvable Streams (2017, background): “Other related topics involving multiple feature sets include multi-view learning [27], [28], transfer learning [29], [30] and incremental attribute learning [31].”
COMPLETER: Incomplete Multi-view Clustering via Contrastive Prediction (2021, methodology): “Second, our method is a deep rather than shallow model [10, 19, 22, 26, 35, 39, 53], thus naturally embracing the capacity of handling complex and large-scale dataset.”
Partial Multi-Modal Sparse Coding via Adaptive Similarity Structure Regularization (2016, background): “However, in real applications, it is often the case that some modalities of the data may suffer from missing information and thus result in partial multi-modality data [15, 9].”
Dual Contrastive Prediction for Incomplete Multi-View Representation Learning (2022, methodology): “First, almost all these methods [19], [20], [22], [23], [25], [26], [27], [28], [29] treat the consistency learning and data recovery as two separate problems.”
A Survey on Multi-View Clustering (2017, unknown): “which is used widely in many works [35], [74], [102].”
Robust Unsupervised Cross-modal Hashing for Multimedia Retrieval (2020, background): “As shown in Figure 1(b), some images or text descriptions do not exist for the query or retrieval databases (such data are called partially paired data [18]), then the pairwise correspondences between them cannot be established.”
Robust Multi-View Clustering With Incomplete Information (2022, methodology): “We compare SURE with 10 multi-view clustering baselines including CCA [12], KCCA [11], DCCA [1], DCCAE [2], BMVC [14], AE(2)-Nets [4], PMVC [51], UEAF [58], DAIMC [16] and EERIMC [17].”
Multi-omic and multi-view clustering algorithms: review and cancer benchmark (2018, methodology): “Finally, PVC [60] performs partial multi-view clustering.”
Multi-Task Multi-View Clustering (2016, unknown): “Manuscript received 29 May 2016; revised 18 Aug. 2016; accepted 19 Aug. 2016.”
Deep Partial Multi-View Learning (2020, background): “However, in practical applications, there are usually incomplete cases for multi-view data [4], [5], [6], [7], [8].”
Multi-view learning overview: Recent progress and new challenges (2017, background): “[23] presented partial multi-view clustering in the case that every view suffers from the missing of some data and results in many partial examples.”
Doubly Aligned Incomplete Multi-view Clustering (2018, methodology): “Fortheevaluation metric, we follow [ Li et al. , 2014 ] , us-ing Normalized Mutual Information (NMI).”