NiGuLa commited on
Commit
9f21b95
1 Parent(s): d3640e8

Create README.md

Browse files
Files changed (1) hide show
  1. README.md +49 -0
README.md ADDED
@@ -0,0 +1,49 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - ru
4
+
5
+ tags:
6
+ - toxic comments classification
7
+
8
+ ---
9
+
10
+ ## General concept of the model
11
+
12
+
13
+ Sensitive topics are such topics that have a high chance of initiating a toxic conversation: homophobia, politics, racism, etc. This dataset uses 18 topics.
14
+
15
+ More details can be found [in this article ](https://www.aclweb.org/anthology/2021.bsnlp-1.4/) presented at the workshop for Balto-Slavic NLP at the EACL-2021 conference.
16
+ This paper presents the first version of this dataset. Here you can see the last version of the dataset which is significantly larger and also properly filtered.
17
+
18
+
19
+ ## Licensing Information
20
+
21
+ [Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License][cc-by-nc-sa].
22
+
23
+ [![CC BY-NC-SA 4.0][cc-by-nc-sa-image]][cc-by-nc-sa]
24
+
25
+ [cc-by-nc-sa]: http://creativecommons.org/licenses/by-nc-sa/4.0/
26
+ [cc-by-nc-sa-image]: https://i.creativecommons.org/l/by-nc-sa/4.0/88x31.png
27
+
28
+ ## Citation
29
+
30
+ If you find this repository helpful, feel free to cite our publication:
31
+
32
+ ```
33
+ @inproceedings{babakov-etal-2021-detecting,
34
+ title = "Detecting Inappropriate Messages on Sensitive Topics that Could Harm a Company{'}s Reputation",
35
+ author = "Babakov, Nikolay and
36
+ Logacheva, Varvara and
37
+ Kozlova, Olga and
38
+ Semenov, Nikita and
39
+ Panchenko, Alexander",
40
+ booktitle = "Proceedings of the 8th Workshop on Balto-Slavic Natural Language Processing",
41
+ month = apr,
42
+ year = "2021",
43
+ address = "Kiyv, Ukraine",
44
+ publisher = "Association for Computational Linguistics",
45
+ url = "https://www.aclweb.org/anthology/2021.bsnlp-1.4",
46
+ pages = "26--36",
47
+ abstract = "Not all topics are equally {``}flammable{''} in terms of toxicity: a calm discussion of turtles or fishing less often fuels inappropriate toxic dialogues than a discussion of politics or sexual minorities. We define a set of sensitive topics that can yield inappropriate and toxic messages and describe the methodology of collecting and labelling a dataset for appropriateness. While toxicity in user-generated data is well-studied, we aim at defining a more fine-grained notion of inappropriateness. The core of inappropriateness is that it can harm the reputation of a speaker. This is different from toxicity in two respects: (i) inappropriateness is topic-related, and (ii) inappropriate message is not toxic but still unacceptable. We collect and release two datasets for Russian: a topic-labelled dataset and an appropriateness-labelled dataset. We also release pre-trained classification models trained on this data.",
48
+ }
49
+ ```