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

In machine learning applications, the reliability of predictions issignificant for assisted decision and risk control. As an effective frameworkto quantify the prediction reliability, conformal prediction (CP) was developedwith the CPKNN (CP with kNN). However, the conventional CPKNN suffers from highvariance and bias and long computational time as the feature dimensionalityincreases. To address these limitations, a new CP framework-conformalprediction with shrunken centroids (CPSC) is proposed. It regularizes the classcentroids to attenuate the irrelevant features and shrink the sample space forpredictions and reliability quantification. To compare CPKNN and CPSC, weemployed them in the classification of 12 categories of alternative herbalmedicine with electronic nose as a case and assessed them in two tasks: 1)offline prediction: the training set was fixed and the accuracy on the testingset was evaluated; 2) online prediction with data augmentation: they filteredunlabeled data to augment the training data based on the prediction reliabilityand the final accuracy of testing set was compared. The result shows that CPSCsignificantly outperformed CPKNN in both two tasks: 1) CPSC reached asignificantly higher accuracy with lower computation cost, and with the samecredibility output, CPSC generally achieves a higher accuracy; 2) the dataaugmentation process with CPSC robustly manifested a statistically significantimprovement in prediction accuracy with different reliability thresholds, andthe augmented data were more balanced in classes. This novel CPSC provideshigher prediction accuracy and better reliability quantification, which can bea reliable assistance in decision support.

CPSC: Conformal prediction with shrunken centroids for efficient  prediction reliability quantification and data augmentation, a case in  alternative herbal medicine classification with electronic nose