ctoraman commited on
Commit
ec27529
1 Parent(s): 0f8ce20

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +37 -0
README.md CHANGED
@@ -1,3 +1,40 @@
1
  ---
 
 
 
 
2
  license: cc-by-nc-sa-4.0
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ language:
3
+ - en
4
+ tags:
5
+ - berturk
6
  license: cc-by-nc-sa-4.0
7
  ---
8
+
9
+ # hate-speech-berturk (base-uncased)
10
+
11
+ Fine-tuned hate speech detection model on Turkish language using Toraman22 v2 dataset published at https://github.com/avaapm/hatespeech
12
+
13
+ Class labels:
14
+
15
+ "0": Neutral
16
+
17
+ "1": Offensive
18
+
19
+ "2": Hate
20
+
21
+ We split 80-20 train-test randomly.
22
+
23
+ 5 epochs train_loss: 0.1571
24
+
25
+ eval_f1 is 0.9094, eval_accuracy is 0.9092
26
+
27
+ ### BibTeX entry and citation info
28
+ ```bibtex
29
+ @InProceedings{toraman2022large,
30
+ author = {Toraman, Cagri and \c{S}ahinu\c{c}, Furkan and Yilmaz, Eyup Halit},
31
+ title = {Large-Scale Hate Speech Detection with Cross-Domain Transfer},
32
+ booktitle = {Proceedings of the Language Resources and Evaluation Conference},
33
+ month = {June},
34
+ year = {2022},
35
+ address = {Marseille, France},
36
+ publisher = {European Language Resources Association},
37
+ pages = {2215--2225},
38
+ url = {https://aclanthology.org/2022.lrec-1.238}
39
+ }
40
+ ```