Next Article in Journal
Deep Learning-Enabled Multitask System for Exercise Recognition and Counting
Next Article in Special Issue
Using Shallow and Deep Learning to Automatically Detect Hate Motivated by Gender and Sexual Orientation on Twitter in Spanish
Previous Article in Journal
Identifying the Characteristics of Virtual Reality Gamification for Complex Educational Topics
Previous Article in Special Issue
Multimodal Hate Speech Detection in Greek Social Media
 
 
Article

Methods for Detoxification of Texts for the Russian Language †

1
Skolkovo Institute of Science and Technology, 121205 Moscow, Russia
2
Mobile TeleSystems (MTS), 109147 Moscow, Russia
*
Author to whom correspondence should be addressed.
This paper is an extended version of our paper published in Dialogue 2021 “Methods for Detoxification of Texts for the Russian Language”.
Academic Editor: Sasa Arsovski
Multimodal Technol. Interact. 2021, 5(9), 54; https://doi.org/10.3390/mti5090054
Received: 3 June 2021 / Revised: 13 August 2021 / Accepted: 20 August 2021 / Published: 4 September 2021
(This article belongs to the Special Issue Hate and Fake: Tackling the Evil in Online Social Media)
We introduce the first study of the automatic detoxification of Russian texts to combat offensive language. This kind of textual style transfer can be used for processing toxic content on social media or for eliminating toxicity in automatically generated texts. While much work has been done for the English language in this field, there are no works on detoxification for the Russian language. We suggest two types of models—an approach based on BERT architecture that performs local corrections and a supervised approach based on a pretrained GPT-2 language model. We compare these methods with several baselines. In addition, we provide the training datasets and describe the evaluation setup and metrics for automatic and manual evaluation. The results show that the tested approaches can be successfully used for detoxification, although there is room for improvement. View Full-Text
Keywords: text style transfer; toxicity detection; detoxification; pretrained models text style transfer; toxicity detection; detoxification; pretrained models
Show Figures

Figure 1

MDPI and ACS Style

Dementieva, D.; Moskovskiy, D.; Logacheva, V.; Dale, D.; Kozlova, O.; Semenov, N.; Panchenko, A. Methods for Detoxification of Texts for the Russian Language. Multimodal Technol. Interact. 2021, 5, 54. https://doi.org/10.3390/mti5090054

AMA Style

Dementieva D, Moskovskiy D, Logacheva V, Dale D, Kozlova O, Semenov N, Panchenko A. Methods for Detoxification of Texts for the Russian Language. Multimodal Technologies and Interaction. 2021; 5(9):54. https://doi.org/10.3390/mti5090054

Chicago/Turabian Style

Dementieva, Daryna, Daniil Moskovskiy, Varvara Logacheva, David Dale, Olga Kozlova, Nikita Semenov, and Alexander Panchenko. 2021. "Methods for Detoxification of Texts for the Russian Language" Multimodal Technologies and Interaction 5, no. 9: 54. https://doi.org/10.3390/mti5090054

Find Other Styles
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Back to TopTop