2026-04-17T17:49:05.774Z http://ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai

ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai:3:10126 2025-06-13T13:58:39Z 3

Online harassment detection on online data science platforms optimized by metaheuristic 2024 http://ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai/record/3/10126 https://doi.org/10.2991/978-94-6463-482-2_9 N. Bacanin M. Kabiljo L. Babic V. Gajic J. Kaljevic M. Dobrojevic Cyberbullying denotes one of the recent pervasive problems, mostly found on social networks, that poses a considerable challenge to keep safe and inclusive environment. It can lead to serious psychological problems for the victim. As one of possible responses, artificial intelligence emerged as a powerful option to identify cases of cyberbullying, and it has garnered considerable attention. This paper suggest using a combination of natural language processing, paired with machine learning XGBoost classifier tuned by an altered variant of the sine cosine metaheuristics to classify and identify the cases of cyberbullying in data collected from a variety of social networks including Kaggle, Twitter and Youtube. The obtained simulation outcomes suggest considerable potential of machine learning models to address this problem. bookPart Atlantis Press 121 136 10.2991/978-94-6463-482-2_9 Proceedings of the 2nd International Conference on Innovation in Information Technology and Business (ICIITB 2024), Chapter in Advances in Computer Science Research