2026-05-04T19:52:50.688Z http://ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai

ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai:3:11577 2025-09-04T22:39:11Z 3

Optimizing AdaBoost by Modified Metaheuristic for Fake News Detection, Chapter in LNNS Lecture Notes in Networks and Systems: ICITI 2024: International Conference on Information Technology and Intelligence, Springer, volume 1342 2025 http://ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai/record/3/11577 https://link.springer.com/chapter/10.1007/978-981-96-5010-1_13 M. Protic L. Jovanovic M. Antonijevic M. Milovanovic M. Zivkovic N. Bacanin A. Jokic Online articles that are deceptive but compelling can seriously affect public discourse, sway public opinion, spark civil unrest, induce fear, and lend credence to hazardous conspiracy theories. Finding these kinds of articles is a difficult task. It’s challenging to manually check every item on the Internet because of the rise in number of articles and the recent release of publicly accessible generative text models. This paper investigates the detection of fake news in article text using natural language processing (NLP). The integration of AdaBoost classification and term frequency inverse document frequency (TF-ID) encoding in addressing this complicated challenge is examined. To ensure optimal performance of the suggested approach, a modified Variable Neighborhood Search (VNS) optimizer is presented to manage the artificial intelligence (AI) algorithm’s parameter modification. The suggested method is tested on a real-world dataset that is openly accessible and tested against eight existing well-known methods. With an accuracy of 90.735%, the gathered results indicate that the suggested model performs better than the others. bookPart Springer, Singapore 153 167 10.1007/978-981-96-5010-1_13 LNNS Lecture Notes in Networks and Systems: ICITI 2024: Proceedings of International Conference on Information Technology and Intelligence, volume 1342