2026-05-04T19:52:44.507Z http://ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai

ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai:3:11574 2025-09-02T17:24:46Z 3

Modifying Metaheuristic Optimizers for Hyperparameter Tuning of Machine Learning Models Tackling Malicious Node Detection in Blockchain Networks, Chapter in AIS Algorithms for Intelligent Systems: ICCI 2024: International Conference on Computational Intelligence, Springer 2025 http://ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai/record/3/11574 https://link.springer.com/chapter/10.1007/978-981-96-4539-8_5 N. Jankovic A. Petrovic L. Jovanovic M. Mihajlovic M. Antonijevic M. Zivkovic N. Bacanin The importance of robust security in blockchain systems extends beyond cryptocurrency to applications in industries like healthcare, where secure data management is crucial. Ensuring early detection of nodes with malicious intent is a critical first step in maintaining the integrity of such networks. This work focuses on addressing this challenge by introducing a machine learning-based solution for detecting and identifying malicious nodes within blockchain networks. The study utilizes the adaptive boosting (AdaBoost) classifier, which is further optimized through hyperparameter tuning using a modified metaheuristic algorithm. The proposed method is compared against several state-of-the-art algorithms and demonstrates strong performance. In the best-case scenario, the system achieved an accuracy of around 87%, highlighting its potential for significantly improving blockchain network security. This research provides a foundation for proactive measures in safeguarding distributed systems from attacks, ensuring their continued reliability across various sectors. bookPart Springer, Singapore 61 73 10.1007/978-981-96-4539-8_5 AIS Algorithms for Intelligent Systems: ICCI 2024: Proceedings of International Conference on Computational Intelligence