Multiple View Clustering Using a Weighted Combination of Exemplar-Based Mixture Models

Tzortzis, Grigorios; Likas, C L

doi:10.1109/tnn.2010.2081999

Cited by 60 publications

(24 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Cluster analysis is a method of unsupervised learning, and in this case the hidden structure that is extracted consists of dividing the input data into separate clusters in a way to optimize a distance measure or a measure of similarity. The choice of the clustering cost function and the optimization algorithm employed to solve the problem determines the resulting clusters (Puzicha et al, 2000; Lashkari and Golland, 2008). …”

Section: Methodsmentioning

confidence: 99%

Adaptive cluster analysis approach for functional localization using magnetoencephalography

Alikhanian¹,

Crawford²,

DeSouza³

et al. 2013

Front. Neurosci.

View full text Add to dashboard Cite

In this paper we propose an agglomerative hierarchical clustering Ward's algorithm in tandem with the Affinity Propagation algorithm to reliably localize active brain regions from magnetoencephalography (MEG) brain signals. Reliable localization of brain areas with MEG has been difficult due to variations in signal strength, and the spatial extent of the reconstructed activity. The proposed approach to resolve this difficulty is based on adaptive clustering on reconstructed beamformer images to find locations that are consistently active across different participants and experimental conditions with high spatial resolution. Using data from a human reaching task, we show that the method allows more accurate and reliable localization from MEG data alone without using functional magnetic resonance imaging (fMRI) or any other imaging techniques.

show abstract

Section: Methodsmentioning

confidence: 99%

Adaptive cluster analysis approach for functional localization using magnetoencephalography

Alikhanian¹,

Crawford²,

DeSouza³

et al. 2013

Front. Neurosci.

View full text Add to dashboard Cite

show abstract

“…• WVCMM: WVCMM [30] is a weighted multi-view clustering called weighted multi-view CMM(convex mixture models) which assume that the data in each view are based on the Gaussian distribution.…”

Section: Comparison With Other Algorithmsmentioning

confidence: 99%

Weighted Multi-view On-Line Competitive Clustering

Zhang

Huang

Wang

et al. 2016

2016 IEEE Second International Conference on Big Data Computing Service and Applications (BigDataService)

View full text Add to dashboard Cite

Multi-view clustering has become a popular clustering technique in recent years due to its ability to analyze data collected from multiple sources or represented by multiple views. In this paper, we propose a novel multi-view clustering approach termed weighted multi-view online competitive clustering (WMLCC). We simultaneously exploit the variable weighting strategy and the online competitive learning in our approach and cast the multi-view clustering problem into an optimization problem. The multi-view clustering result can be obtained by optimizing a new objective function. We conduct experiments on two real-world multi-view datasets. Experimental results demonstrate the effectiveness and efficiency of our approach.

show abstract

“…The traditional way is to build a probabilistic model for each feature type, and then estimate a mixture of them [3,27]. Alternatively, relying on kernel combination, one can use each feature type to compute a similarity kernel matrix for a weighted sum [16,39].…”

Section: Related Workmentioning

confidence: 99%

Multi-feature Spectral Clustering with Minimax Optimization

Wang

Weng

Yuan

2014

2014 IEEE Conference on Computer Vision and Pattern Recognition

View full text Add to dashboard Cite

In this paper, we propose a novel formulation for multifeature clustering using minimax optimization. To find a consensus clustering result that is agreeable to all feature modalities, our objective is to find a universal feature embedding, which not only fits each individual feature modality well, but also unifies different feature modalities by minimizing their pairwise disagreements. The loss function consists of both (1) unary embedding cost for each modality, and (2) pairwise disagreement cost for each pair of modalities, with weighting parameters automatically selected to maximize the loss. By performing minimax optimization, we can minimize the loss for the worst case with maximum disagreements, thus can better reconcile different feature modalities. To solve the minimax optimization, an iterative solution is proposed to update the universal embedding, individual embedding, and fusion weights, separately. Our minimax optimization has only one global parameter. The superior results on various multi-feature clustering tasks validate the effectiveness of our approach when compared with the state-of-the-art methods.

show abstract

Multiple View Clustering Using a Weighted Combination of Exemplar-Based Mixture Models

Cited by 60 publications

References 20 publications

Adaptive cluster analysis approach for functional localization using magnetoencephalography

Adaptive cluster analysis approach for functional localization using magnetoencephalography

Weighted Multi-view On-Line Competitive Clustering

Multi-feature Spectral Clustering with Minimax Optimization

Contact Info

Product

Resources

About