SemEval 2020 paper on offensive language detection (OffensEval 2020)

[Ted Pedersen]

I participated in the offensive language detection task in SemEval
2020. I worked on my own and used a really simple method (logistic
regression) which I suspected would do fairly pooly, but my goal
really was to just get into an analysis of the results and data
post-evaluation. The paper that resulted from all that is now
available on arxiv. It will eventually appear in the SemEval 2020
proceedings, although that will not be until December since Coling and
SemEval have been postponed until then.

https://arxiv.org/abs/2007.12946

As always, questions, comments, suggestions are most welcome.
---
Ted Pedersen
http://www.d.umn.edu/~tpederse