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

Feature selection is a dimensionality reduction technique that selects asubset of representative features from high dimensional data by eliminatingirrelevant and redundant features. Recently, feature selection combined withsparse learning has attracted significant attention due to its outstandingperformance compared with traditional feature selection methods that ignorescorrelation between features. These works first map data onto a low-dimensionalsubspace and then select features by posing a sparsity constraint on thetransformation matrix. However, they are restricted by design to linear datatransformation, a potential drawback given that the underlying correlationstructures of data are often non-linear. To leverage a more sophisticatedembedding, we propose an autoencoder-based unsupervised feature selectionapproach that leverages a single-layer autoencoder for a joint framework offeature selection and manifold learning. More specifically, we enforce columnsparsity on the weight matrix connecting the input layer and the hidden layer,as in previous work. Additionally, we include spectral graph analysis on theprojected data into the learning process to achieve local data geometrypreservation from the original data space to the low-dimensional feature space.Extensive experiments are conducted on image, audio, text, and biological data.The promising experimental results validate the superiority of the proposedmethod.

Graph Autoencoder-Based Unsupervised Feature Selection with Broad and Local Data Structure Preservation