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

Mutual Information (MI) based feature selection makes use of MI to evaluateeach feature and eventually shortlists a relevant feature subset, in order toaddress issues associated with high-dimensional datasets. Despite theeffectiveness of MI in feature selection, we notice that many state-of-the-artalgorithms disregard the so-called unique relevance (UR) of features, andarrive at a suboptimal selected feature subset which contains a non-negligiblenumber of redundant features. We point out that the heart of the problem isthat all these MIBFS algorithms follow the criterion of Maximize Relevance withMinimum Redundancy (MRwMR), which does not explicitly target UR. This motivatesus to augment the existing criterion with the objective of boosting uniquerelevance (BUR), leading to a new criterion called MRwMR-BUR. Depending on thetask being addressed, MRwMR-BUR has two variants, termed MRwMR-BUR-KSG andMRwMR-BUR-CLF, which estimate UR differently. MRwMR-BUR-KSG estimates UR via anearest-neighbor based approach called the KSG estimator and is designed forthree major tasks: (i) Classification Performance. (ii) FeatureInterpretability. (iii) Classifier Generalization. MRwMR-BUR-CLF estimates URvia a classifier based approach. It adapts UR to different classifiers, furtherimproving the competitiveness of MRwMR-BUR for classification performanceoriented tasks. The performance of both MRwMR-BUR-KSG and MRwMR-BUR-CLF isvalidated via experiments using six public datasets and three popularclassifiers. Specifically, as compared to MRwMR, the proposed MRwMR-BUR-KSGimproves the test accuracy by 2% - 3% with 25% - 30% fewer features beingselected, without increasing the algorithm complexity. MRwMR-BUR-CLF furtherimproves the classification performance by 3.8%- 5.5% (relative to MRwMR), andit also outperforms three popular classifier dependent feature selectionmethods.

Improving Mutual Information based Feature Selection by Boosting Unique Relevance