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

Traditional hand-crafted linguistically-informed features have often beenused for distinguishing between translated and original non-translated texts.By contrast, to date, neural architectures without manual feature engineeringhave been less explored for this task. In this work, we (i) compare thetraditional feature-engineering-based approach to the feature-learning-basedone and (ii) analyse the neural architectures in order to investigate how wellthe hand-crafted features explain the variance in the neural models'predictions. We use pre-trained neural word embeddings, as well as severalend-to-end neural architectures in both monolingual and multilingual settingsand compare them to feature-engineering-based SVM classifiers. We show that (i)neural architectures outperform other approaches by more than 20 accuracypoints, with the BERT-based model performing the best in both the monolingualand multilingual settings; (ii) while many individual hand-craftedtranslationese features correlate with neural model predictions, featureimportance analysis shows that the most important features for neural andclassical architectures differ; and (iii) our multilingual experiments provideempirical evidence for translationese universals across languages.

Comparing Feature-Engineering and Feature-Learning Approaches for Multilingual Translationese Classification