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

Performing classification on noisy, crowdsourced image datasets can provechallenging even for the best neural networks. Two issues which complicate theproblem on such datasets are class imbalance and ground-truth uncertainty inlabeling. The AL-ALL and AL-PUB datasets - consisting of tightly cropped,individual characters from images of ancient Greek papyri - are stronglyaffected by both issues. The application of ensemble modeling to such datasetscan help identify images where the ground-truth is questionable and quantifythe trustworthiness of those samples. As such, we apply stacked generalizationconsisting of nearly identical ResNets with different loss functions: oneutilizing sparse cross-entropy (CXE) and the other Kullback-Liebler Divergence(KLD). Both networks use labels drawn from a crowd-sourced consensus. Thisconsensus is derived from a Normalized Distribution of Annotations (NDA) basedon all annotations for a given character in the dataset. For the secondnetwork, the KLD is calculated with respect to the NDA. For our ensemble model,we apply a k-nearest neighbors model to the outputs of the CXE and KLDnetworks. Individually, the ResNet models have approximately 93% accuracy,while the ensemble model achieves an accuracy of > 95%, increasing theclassification trustworthiness. We also perform an analysis of the Shannonentropy of the various models' output distributions to measure classificationuncertainty. Our results suggest that entropy is useful for predicting modelmisclassifications.

Incorporating Crowdsourced Annotator Distributions into Ensemble Modeling to Improve Classification Trustworthiness for Ancient Greek Papyri