Filtering information distributed in social networks and identifying threats
Keywords:
artificial intelligence, YouTube API, information security, inappropriate content, threatening comments, comment filtering, disinformation, incitement to hatred, automated analysis, Python, NLP, Perspective APIAbstract
The article discusses the possibilities of using artificial intelligence technologies to identify and filter content containing obscene language or threats distributed on social networks, in particular on the YouTube platform. It considers the extraction of comments through the YouTube API, and then the analysis and practical application of these comments using artificial intelligence. The significance, limitations and proposed solutions of this approach to ensuring information security in our republic are substantiated.
References
1. Russell S., Norvig P. Sun’iy intellekt: zamonaviy yondashuv. – 3-nashr. – Pearson Education, 2016.
2. Bird S., Klein E., Loper E. Natural Language Processing with Python. – O’Reilly Media, 2009.
3. Google Developers. Perspective API hujjatlari. – https://developers.perspectiveapi.com
4. YouTube Data API v3. – Rasmiy hujjat. – https://developers.google.com/youtube/v3
5. Hovy E., Lavid J. Towards a “Science” of Natural Language Processing. // Computer Speech and Language, 2010.
6. Axborot xavfsizligi to‘g‘risidagi O‘zbekiston Respublikasi Qonuni. – 2020-yil, 11-dekabr. – www.lex.uz
7. Sherzodbek K., Yusupov M. O‘zbek tilida NLP yondashuvlari: imkoniyatlar va muammolar. // TATU ilmiy axborotnomasi, 2022.
8. Luhn H.P. The Automatic Creation of Literature Abstracts. // IBM Journal of Research and Development, 1958.
