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 recent years, there has been increased interest in building predictivemodels that harness natural language processing and machine learning techniquesto detect emotions from various text sources, including social media posts,micro-blogs or news articles. Yet, deployment of such models in real-worldsentiment and emotion applications faces challenges, in particular poorout-of-domain generalizability. This is likely due to domain-specificdifferences (e.g., topics, communicative goals, and annotation schemes) thatmake transfer between different models of emotion recognition difficult. Inthis work we propose approaches for text-based emotion detection that leveragetransformer models (BERT and RoBERTa) in combination with Bidirectional LongShort-Term Memory (BiLSTM) networks trained on a comprehensive set ofpsycholinguistic features. First, we evaluate the performance of our modelswithin-domain on two benchmark datasets: GoEmotion and ISEAR. Second, weconduct transfer learning experiments on six datasets from the Unified EmotionDataset to evaluate their out-of-domain robustness. We find that the proposedhybrid models improve the ability to generalize to out-of-distribution datacompared to a standard transformer-based approach. Moreover, we observe thatthese models perform competitively on in-domain data.

Improving the Generalizability of Text-Based Emotion Detection by Leveraging Transformers with Psycholinguistic Features