Options
Автоматичне розпізнавання Телеграм-каналів з маніпулятивною риторикою на основі лінгвістичних та екстралінгвістичних критеріїв
Issue Date :
2024
Author(s) :
Кулдошинова Марія
Academic supervisor(s)/editor(s) :
Костіков Микола Павлович
Abstract :
In everyday communication mediums, such as messaging apps and social networks, users frequently encounter manipulative statements that may be perceived as truthful when within the information flow. In the context of the Russian-Ukrainian war, this issue is extremely relevant and raises the issue of enhancing the information security level of Ukrainian citizens.
This study was conducted with the aim of creating software that automates the process of recognizing manipulative rhetoric in news Telegram channels.
The objects of the research are the linguistic and extralinguistic markers of manipulative rhetoric, and the subject is a computer model for their identification. The goal of the study is to create an algorithm for analyzing news texts and detecting signs of manipulation. To achieve this, the ethical standards of journalism and existing methods for recognizing disinformation in texts are analyzed, and markers for the automatic evaluation of news publications are determined based on the collected data.
The methodological base includes methods of automatic text analysis, computer modeling, morphological analysis, extralinguistic analysis, and statistical analysis. Thus, linguistic and technical approaches are combined to identify manipulative rhetoric.
As a result of the study, software and its implementation in a Telegram bot were developed. Practical testing confirmed the effectiveness of the tool. The results indicate significant potential of the methodology for enhancing information security and supporting media literacy among the population in the face of constant information attacks.
Thus, the conducted research makes an important contribution to the development of tools for combating hostile information influence in wartime, offering a novel approach and an accessible way to verify media content.
This study was conducted with the aim of creating software that automates the process of recognizing manipulative rhetoric in news Telegram channels.
The objects of the research are the linguistic and extralinguistic markers of manipulative rhetoric, and the subject is a computer model for their identification. The goal of the study is to create an algorithm for analyzing news texts and detecting signs of manipulation. To achieve this, the ethical standards of journalism and existing methods for recognizing disinformation in texts are analyzed, and markers for the automatic evaluation of news publications are determined based on the collected data.
The methodological base includes methods of automatic text analysis, computer modeling, morphological analysis, extralinguistic analysis, and statistical analysis. Thus, linguistic and technical approaches are combined to identify manipulative rhetoric.
As a result of the study, software and its implementation in a Telegram bot were developed. Practical testing confirmed the effectiveness of the tool. The results indicate significant potential of the methodology for enhancing information security and supporting media literacy among the population in the face of constant information attacks.
Thus, the conducted research makes an important contribution to the development of tools for combating hostile information influence in wartime, offering a novel approach and an accessible way to verify media content.
Bibliographic description :
Кулдошинова М. Автоматичне розпізнавання Телеграм-каналів з маніпулятивною риторикою на основі лінгвістичних та екстралінгвістичних критеріїв : кваліфікаційна робота бакалавра : 035.10 Філологія (прикладна лінгвістика) / наук. кер. М. Костіков. Київ, 2024. 71 с.
File(s) :
Loading...
Format
Adobe PDF
Size :
3.32 MB
Checksum :
(MD5):3ea478cbe4f8809fa7e8562e3a8a8430
This work is distributed under the Creative Commons license CC BY-NC