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

Detecting controversy in general web pages is a daunting task, butincreasingly essential to efficiently moderate discussions and effectivelyfilter problematic content. Unfortunately, controversies occur across manytopics and domains, with great changes over time. This paper investigatesneural classifiers as a more robust methodology for controversy detection ingeneral web pages. Current models have often cast controversy detection ongeneral web pages as Wikipedia linking, or exact lexical matching tasks. Thediverse and changing nature of controversies suggest that semantic approachesare better able to detect controversy. We train neural networks that cancapture semantic information from texts using weak signal data. By leveragingthe semantic properties of word embeddings we robustly improve on existingcontroversy detection methods. To evaluate model stability over time and tounseen topics, we asses model performance under varying training conditions totest cross-temporal, cross-topic, cross-domain performance and annotatorcongruence. In doing so, we demonstrate that weak-signal based neuralapproaches are closer to human estimates of controversy and are more robust tothe inherent variability of controversies.

Improved and Robust Controversy Detection in General Web Pages Using Semantic Approaches under Large Scale Conditions