Automatic Semantic Role Labelling (SRL) in Swedish

Yang Buhr, Lucy
Göteborgs universitet/Institutionen för data- och informationsteknikswe
University of Gothenburg/Department of Computer Science and Engineeringeng
2023-10-05T14:49:42Z
2023-10-05T14:49:42Z
2023-10-05
In this paper, using deep learning networks, the first end-to-end semantic role labelling model (SRL) has been developed for Swedish texts. This Swedish SRL model can, with a given Swedish sentence, perform trigger identification, frame classification and argument extraction tasks automatically in a series. Using the semantic annotation examples from the Swedish FrameNet corpus (SweFN), we experiment with two transfer learning approaches. We show that the Swedish SRL based on a pre-trained English SRL can speed up the training, meanwhile the multilingual model (mT5) based model has better performance when the possible frames are unknown. Through extensive empirical analysis of the model performance, we point out the major factors that can further improve the results.en
https://hdl.handle.net/2077/78681
engen
Technology
Semantic role labellingen
SRLen
transfer learningen
multi-task learningen
multilingual transferen
T5en
mT5en
Automatic Semantic Role Labelling (SRL) in Swedishen
text
Student essay
H2

Files

Original bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
CSE 23-08 LYB.pdf
Size:
1.08 MB
Format:
Adobe Portable Document Format
Description:

License bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
license.txt
Size:
876 B
Format:
Item-specific license agreed upon to submission
Description:

Collections