UKP ASPECT Corpus
Description
The UKP ASPECT Corpus includes 3,595 sentence pairs over 28 controversial topics. The sentences were crawled from a large web crawl and identified as arguments for a given topic using the ArgumenText system. The sampling and matching of the sentence pairs is described in the paper. Then, the argument similarity annotation was done via crowdsourcing. Each crowd worker could choose from four annotation options (the exact guidelines are provided in the Appendix of the paper).
Related Resources
- Is supplement to: https://aclanthology.org/P19-1054
Reference in TUbiblio
Collections
-
Argument Mining [19]
The following license files are associated with this item:
Except where otherwise noted, this item's license is described as CC BY-NC 3.0