Hugging Face's logo

Datasets:
FrancophonIA
/
CONAN

Tasks:
Text Classification
Modalities:
Text
Formats:
csv
Sub-tasks:
language-modeling
sentiment-classification
sentiment-scoring
Languages:
English
French
Italian
Size:
10K - 100K
Libraries:
Datasets
pandas
Dataset card Data Studio Files
xet
Community
CONAN
File size: 2,802 Bytes 
f89eb07
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
c1a7887
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
 
---
language:
- eng
- fra
- ita
multilinguality:
  - multilingual
task_categories:
- text-classification
task_ids:
- language-modeling
- sentiment-classification
- sentiment-scoring
size_categories:
- 10K<n<100K
---

> [!NOTE]
> Dataset origin: https://github.com/marcoguerini/CONAN



# Dataset Card for [CONAN]


## Dataset Description

- **Repository:** https://github.com/marcoguerini/CONAN
- **Paper:** https://aclanthology.org/P19-1271/
- **Point of Contact:** ychung@fbk.eu, tekiroglu@fbk.eu, guerini@fbk.eu, elizaveta.kuzmenko@studenti.unitn.it

### Dataset Summary

CONAN is a multilingual and expert-based dataset of hate speech/counter-narrative pairs for English, French and Italian.  
The dataset consists of 4,078 pairs over the 3 languages. Together with the data we also provide 3 types of metadata: expert demographics, hate speech sub-topic and counter-narrative type. The dataset is augmented through translation (from Italian/French to English) and paraphrasing, which brought the total number of pairs to 14.988.  
(*)The original number was 15.024 but after post-hoc analysis, we deleted 9 original pairs (36 pairs including augmented ones) because they did not meet the required standard.

### Supported Tasks and Leaderboards

Text classification

### Languages
English, French and Italian



## Dataset Structure

### Data Instances

Zntry starts with an ID, followed by a pair of hate speech/counter-narrative and the metadata (hate speech type, hate speech sub-topic, counter-narrative type, and demographics).  
ID indicates language, hate speech type, hate speech sub-topic, unique hate speech count, counter-narrative count, and augmentation type (if any).


### Data Fields

ID example:  

FRT1ST0014HS0015CN001238T1  

FR|T1|ST0014|HS0015|CN001238|T1  

Language | HS Type | HS SubTopic | HS ID | CN Count | augmentation type (P1/P2/T1)  

(P1: paraphrase 1 / P2: paraphrase 2 / T1: translation 1)


### Data Splits

No splits provided in the dataset


## Additional Information

### Licensing Information

These resources can be used for research purposes. Please cite the corresponding publication if you use it.

### Citation Information

> @inproceedings{chung-etal-2019-conan,
    title = "{CONAN} - {CO}unter {NA}rratives through Nichesourcing: a Multilingual Dataset of Responses to Fight Online Hate Speech",
    author = "Chung, Yi-Ling and Kuzmenko, Elizaveta and Tekiroglu, Serra Sinem and Guerini, Marco",
    booktitle = "Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics",
    month = jul,
    year = "2019",
    address = "Florence, Italy",
    publisher = "Association for Computational Linguistics",
    url = "https://www.aclweb.org/anthology/P19-1271",
    doi = "10.18653/v1/P19-1271",
    pages = "2819--2829"
}