We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:


References & Citations

DBLP - CS Bibliography


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Computation and Language

Title: Emotion Stimulus Detection in German News Headlines

Abstract: Emotion stimulus extraction is a fine-grained subtask of emotion analysis that focuses on identifying the description of the cause behind an emotion expression from a text passage (e.g., in the sentence "I am happy that I passed my exam" the phrase "passed my exam" corresponds to the stimulus.). Previous work mainly focused on Mandarin and English, with no resources or models for German. We fill this research gap by developing a corpus of 2006 German news headlines annotated with emotions and 811 instances with annotations of stimulus phrases. Given that such corpus creation efforts are time-consuming and expensive, we additionally work on an approach for projecting the existing English GoodNewsEveryone (GNE) corpus to a machine-translated German version. We compare the performance of a conditional random field (CRF) model (trained monolingually on German and cross-lingually via projection) with a multilingual XLM-RoBERTa (XLM-R) model. Our results show that training with the German corpus achieves higher F1 scores than projection. Experiments with XLM-R outperform their respective CRF counterparts.
Comments: accepted at KONVENS 2021
Subjects: Computation and Language (cs.CL)
Cite as: arXiv:2107.12920 [cs.CL]
  (or arXiv:2107.12920v2 [cs.CL] for this version)

Submission history

From: Roman Klinger [view email]
[v1] Tue, 27 Jul 2021 16:22:04 GMT (30kb,D)
[v2] Wed, 28 Jul 2021 07:21:13 GMT (30kb,D)

Link back to: arXiv, form interface, contact.