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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

With the abundance of machine learning methods available and the temptationof using them all in an ensemble method, having a model-agnostic method offeature selection is incredibly alluring. Principal component analysis wasdeveloped in 1901 and has been a strong contender in this role since, but inthe end is an unsupervised method. It offers no guarantee that the featuresthat are selected have good predictive power because it does not know what isbeing predicted. To this end, Peng et al. developed the minimumredundancy-maximum relevance (mRMR) method in 2005. It uses the mutualinformation not only between predictors but also includes the mutualinformation with the response in its calculation. Estimating mutual informationand entropy tend to be expensive and problematic endeavors, which leads toexcessive processing times even for dataset that is approximately 750 by 750 ina Leave-One-Subject-Out jackknife situation. To remedy this, we use a methodfrom 2012 called Distance Correlation Sure Independence Screening (DC-SIS)which uses the distance correlation measure of Sz\'ekely et al. to selectfeatures that have the greatest dependence with the response. We show that thismethod produces statistically indistinguishable results to the mRMR selectionmethod on Parkinson's Disease vocal diagnosis data 90 times faster.

Distance Correlation Sure Independence Screening for Accelerated Feature Selection in Parkinson's Disease Vocal Data