English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Clustering algorithms have significantly improved along with Deep NeuralNetworks which provide effective representation of data. Existing methods arebuilt upon deep autoencoder and self-training process that leverages thedistribution of cluster assignments of samples. However, as the fundamentalobjective of the autoencoder is focused on efficient data reconstruction, thelearnt space may be sub-optimal for clustering. Moreover, it requires highlyeffective codes (i.e., representation) of data, otherwise the initial clustercenters often cause stability issues during self-training. Manystate-of-the-art clustering algorithms use convolution operation to extractefficient codes but their applications are limited to image data. In thisregard, we propose an end-to-end deep clustering algorithm, i.e., Very CompactClusters (VCC). VCC takes advantage of distributions of local relationships ofsamples near the boundary of clusters, so that they can be properly separatedand pulled to cluster centers to form compact clusters. Experimental results onvarious datasets illustrate that our proposed approach achieves competitiveclustering performance against most of the state-of-the-art clustering methodsfor both image and non-image data, and its results can be easily qualitativelyseen in the learnt low-dimensional space.

Separating Boundary Points via Structural Regularization for Very Compact Clusters