Els Lefever

3 paper cards in Mova's library · 2016–2024

Els Lefever wrote 3 of the papers Mova reads, in 2016–2024, most of them on Akkadian and cuneiform NLP, Part-of-speech tagging; often with Cynthia Van Hee, Véronique Hoste, Gustav Ryberg Smidt.

The profile is the library's match of the name; two people with one name are not always told apart.

Papers

At the Crossroad of Cuneiform and NLP: Challenges for Fine-grained Part-of-speech Tagging (2024)
The authors prepare a corpus of 121 Old Babylonian Akkadian letters from Sippar, annotated by three specialists with PoS tags, citation forms and morphemes in the ORACC style.
We Usually Don't Like Going to the Dentist: Using Common Sense to Detect Irony on Twitter (2018)
Most ironic tweets contrast a literal positive evaluation with a situation everyone knows to be bad («working a double on New Year's Eve», «my car won't start»).
Monday mornings are my fave :) #not Exploring the Automatic Recognition of Irony in English tweets (2016)
3,000 English tweets collected by #irony, #sarcasm and #not were all labelled by hand with a fine-grained scheme instead of trusting the hashtags: ironic by a polarity clash, other verbal irony, situational irony, or not ironic (2,396 ironic, 604 not; kappa 0.72), with harsh (mocking) irony flagged, and below tweet…