Modern Approaches to Detect and Classify Comment Toxicity Using Neural Networks

The growth of popularity of online platforms which allow users to communicate with each other, share opinions about various events, and leave comments boosted the development of natural language processing algorithms. Tens of millions of messages per day are published by users of a particular social...

Full description

Saved in:
Bibliographic Details
Main Author: Sergey V. Morzhov
Format: Article
Language:English
Published: Yaroslavl State University 2020-03-01
Series:Моделирование и анализ информационных систем
Subjects:
Online Access:https://www.mais-journal.ru/jour/article/view/1287
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849338870015983616
author Sergey V. Morzhov
author_facet Sergey V. Morzhov
author_sort Sergey V. Morzhov
collection DOAJ
description The growth of popularity of online platforms which allow users to communicate with each other, share opinions about various events, and leave comments boosted the development of natural language processing algorithms. Tens of millions of messages per day are published by users of a particular social network need to be analyzed in real time for moderation in order to prevent the spread of various illegal or offensive information, threats and other types of toxic comments. Of course, such a large amount of information can be processed quite quickly only automatically. that is why there is a need to and a way to teach computers to “understand” a text written by humans. It is a non-trivial task even if the word “understand” here means only “to classify”. the rapid evolution of machine learning technologies has led to ubiquitous implementation of new algorithms. A lot of tasks, which for many years were considered almost impossible to solve, are now quite successfully solved using deep learning technologies. this article considers algorithms built using deep learning technologies and neural networks which can successfully solve the problem of detection and classification of toxic comments. In addition, the article presents the results of the developed algorithms, as well as the results of the ensemble of all considered algorithms on a large training set collected and tagged by Google and Jigsaw.
format Article
id doaj-art-d60b14792c03416b96bcef394323c457
institution Kabale University
issn 1818-1015
2313-5417
language English
publishDate 2020-03-01
publisher Yaroslavl State University
record_format Article
series Моделирование и анализ информационных систем
spelling doaj-art-d60b14792c03416b96bcef394323c4572025-08-20T03:44:17ZengYaroslavl State UniversityМоделирование и анализ информационных систем1818-10152313-54172020-03-01271486110.18255/1818-1015-2020-1-48-61958Modern Approaches to Detect and Classify Comment Toxicity Using Neural NetworksSergey V. Morzhov0P. G. Demidov Yaroslavl State UniversityThe growth of popularity of online platforms which allow users to communicate with each other, share opinions about various events, and leave comments boosted the development of natural language processing algorithms. Tens of millions of messages per day are published by users of a particular social network need to be analyzed in real time for moderation in order to prevent the spread of various illegal or offensive information, threats and other types of toxic comments. Of course, such a large amount of information can be processed quite quickly only automatically. that is why there is a need to and a way to teach computers to “understand” a text written by humans. It is a non-trivial task even if the word “understand” here means only “to classify”. the rapid evolution of machine learning technologies has led to ubiquitous implementation of new algorithms. A lot of tasks, which for many years were considered almost impossible to solve, are now quite successfully solved using deep learning technologies. this article considers algorithms built using deep learning technologies and neural networks which can successfully solve the problem of detection and classification of toxic comments. In addition, the article presents the results of the developed algorithms, as well as the results of the ensemble of all considered algorithms on a large training set collected and tagged by Google and Jigsaw.https://www.mais-journal.ru/jour/article/view/1287toxicitynatural language processingnlpdeep learningword embeddingglovefasttextrecurrent neural networksconvolutional neural networkscnnlstmgru
spellingShingle Sergey V. Morzhov
Modern Approaches to Detect and Classify Comment Toxicity Using Neural Networks
Моделирование и анализ информационных систем
toxicity
natural language processing
nlp
deep learning
word embedding
glove
fasttext
recurrent neural networks
convolutional neural networks
cnn
lstm
gru
title Modern Approaches to Detect and Classify Comment Toxicity Using Neural Networks
title_full Modern Approaches to Detect and Classify Comment Toxicity Using Neural Networks
title_fullStr Modern Approaches to Detect and Classify Comment Toxicity Using Neural Networks
title_full_unstemmed Modern Approaches to Detect and Classify Comment Toxicity Using Neural Networks
title_short Modern Approaches to Detect and Classify Comment Toxicity Using Neural Networks
title_sort modern approaches to detect and classify comment toxicity using neural networks
topic toxicity
natural language processing
nlp
deep learning
word embedding
glove
fasttext
recurrent neural networks
convolutional neural networks
cnn
lstm
gru
url https://www.mais-journal.ru/jour/article/view/1287
work_keys_str_mv AT sergeyvmorzhov modernapproachestodetectandclassifycommenttoxicityusingneuralnetworks