Datasets:

Modalities:
Text
ArXiv:
Libraries:
Datasets
leondz commited on
Commit
2353a09
1 Parent(s): b2b906e

remove incorrect content

Browse files
Files changed (1) hide show
  1. README.md +14 -16
README.md CHANGED
@@ -57,28 +57,26 @@ extra_gated_prompt: "Warning: this repository contains harmful content (abusive
57
  - **Repository:**
58
  - **Paper:** [https://aclanthology.org/2020.semeval-1.188/](https://aclanthology.org/2020.semeval-1.188/)
59
  - **Point of Contact:** [Leon Derczynski](https://github.com/leondz)
60
- - **Size of downloaded dataset files:** 769.21 KiB
61
- - **Size of the generated dataset:** 1.06 MiB
62
- - **Total amount of disk used:** 1.85 MiB
63
 
64
  ### Dataset Summary
65
 
66
- This is an abusive/offensive language detection dataset for Albanian. The data is formatted
67
- following the OffensEval convention, with three tasks:
68
 
69
- * Subtask A: Offensive (OFF) or not (NOT)
70
- * Subtask B: Untargeted (UNT) or targeted insult (TIN)
71
- * Subtask C: Type of target: individual (IND), group (GRP), or other (OTH)
 
 
72
 
73
- Notes on the above:
 
 
74
 
75
- * The subtask A field should always be filled.
76
- * The subtask B field should only be filled if there's "offensive" (OFF) in A.
77
- * The subtask C field should only be filled if there's "targeted" (TIN) in B.
78
 
79
- The dataset name is a backronym, also standing for "Spoken Hate in the Albanian Jargon"
80
-
81
- See the paper [https://arxiv.org/abs/2107.13592](https://arxiv.org/abs/2107.13592) for full details.
82
 
83
  ### Supported Tasks and Leaderboards
84
 
@@ -86,7 +84,7 @@ See the paper [https://arxiv.org/abs/2107.13592](https://arxiv.org/abs/2107.1359
86
 
87
  ### Languages
88
 
89
- Albanian (`bcp47:sq-AL`)
90
 
91
  ## Dataset Structure
92
 
 
57
  - **Repository:**
58
  - **Paper:** [https://aclanthology.org/2020.semeval-1.188/](https://aclanthology.org/2020.semeval-1.188/)
59
  - **Point of Contact:** [Leon Derczynski](https://github.com/leondz)
 
 
 
60
 
61
  ### Dataset Summary
62
 
63
+ OffensEval 2020 features a multilingual dataset with five languages. The languages included in OffensEval 2020 are:
 
64
 
65
+ * Arabic
66
+ * Danish
67
+ * English
68
+ * Greek
69
+ * Turkish
70
 
71
+ The annotation follows the hierarchical tagset proposed in the Offensive Language Identification Dataset (OLID) and used in OffensEval 2019.
72
+ In this taxonomy we break down offensive content into the following three sub-tasks taking the type and target of offensive content into account.
73
+ The following sub-tasks were organized:
74
 
75
+ * Sub-task A - Offensive language identification;
76
+ * Sub-task B - Automatic categorization of offense types;
77
+ * Sub-task C - Offense target identification.
78
 
79
+ English training data is omitted so needs to be collected otherwise (see [https://zenodo.org/record/3950379#.XxZ-aFVKipp](https://zenodo.org/record/3950379#.XxZ-aFVKipp))
 
 
80
 
81
  ### Supported Tasks and Leaderboards
82
 
 
84
 
85
  ### Languages
86
 
87
+ Five are covered: bcp47 `ar;da;en;gr;tr`
88
 
89
  ## Dataset Structure
90