Article · Вестник Алматинского университета энергетики и связи · September 30, 2023

CREATING HATE SPEECH DETECTION MODEL BY USING MACHINE LEARNING METHODS

Batyrkhan Sultanovich Omarov, Elvira Adylbekova, Abay Tursynbayev, Aigerim Toktarova

Abstract

This research paper describes the using machine learning algorithms to pre-identify profanity in online content due to concerns about the growing use of profanity on social networks such as Twitter. In this article, comparative studies were carried out to develop special software using machine learning methods for effective detection and monitoring of hate speech in order to minimize the negative impact of hate speech on people and communities. The scientific novelty of this research work an algorithm defined that provides the most effective accuracy of automatic detection of obscene words among machine learning algorithms. The result of the experiment, the effectiveness of the above methods for detecting offensive words evaluated on Twitter. The purpose of this article is to compile a data set of hate speech collected from various sources and approved by experts to ensure that the obscene words detection model works effectively. The practical significance of this research work, using machine learning methods, a comprehensive method for detecting profanity on Twitter proposed, a detection model and software for recognizing complex elements of profanity in online comments are also provided.

Hate Speech and Cyberbullying DetectionText and Document Classification TechnologiesSwearing, Euphemism, Multilingualismhate speech detectionprofanity detectionTwitter

CREATING HATE SPEECH DETECTION MODEL BY USING MACHINE LEARNING METHODS

Abstract

More by Elvira Adylbekova