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

As growing usage of social media websites in the recent decades, the amountof news articles spreading online rapidly, resulting in an unprecedented scaleof potentially fraudulent information. Although a plenty of studies haveapplied the supervised machine learning approaches to detect such content, thelack of gold standard training data has hindered the development. Analysing thesingle data format, either fake text description or fake image, is themainstream direction for the current research. However, the misinformation inreal-world scenario is commonly formed as a text-image pair where the newsarticle/news title is described as text content, and usually followed by therelated image. Given the strong ability of learning features without labelleddata, contrastive learning, as a self-learning approach, has emerged andachieved success on the computer vision. In this paper, our goal is to explorethe constrastive learning in the domain of misinformation identification. Wedeveloped a self-learning model and carried out the comprehensive experimentson a public data set named COSMOS. Comparing to the baseline classifier, ourmodel shows the superior performance of non-matched image-text pair detection(approximately 10%) when the training data is insufficient. In addition, weobserved the stability for contrsative learning and suggested the use of itoffers large reductions in the number of training data, whilst maintainingcomparable classification results.

Harnessing the Power of Text-image Contrastive Models for Automatic Detection of Online Misinformation