2026-05-20T15:32:09.376Z http://ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai

ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai:3:9508 2024-12-03T20:10:47Z 3

Tackling IoT Security Challenge by Metaheuristics Tuned Extreme Learning Machine, Chapter in LNNS Lecture Notes in Networks and Systems: ICoISS 2023: International Conference on Intelligent Sustainable Systems, Springer, volume 665 2023 http://ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai/record/3/9508 https://link.springer.com/chapter/10.1007/978-981-99-1726-6_39 L. Jovanovic M. Gajevic M. Dobrojevic N. Budimirovic N. Bacanin M. Zivkovic The Internet of Things (IoT) brings together the real and online world to enhance services such as reducing traffic congestion, improving health care, and increasing customer service speed through quicker communication and better analysis. The manufacturing industry’s embrace of new technologies that include wireless sensor networks, embedded devices, cloud systems, and big data has enabled the emergence of Industry 4.0. The Fourth Industrial Revolution has resulted in a technologically improved globalized world in which modern technologies may directly manage the manufacturing industry’s machinery, facilities, plants, and infrastructure, as well as affect intelligent procedural and strategic decisions. The emphasis of this study is on the implementation of a novel metaheuristic optimization method. It is used for feature selection (FS), hyperparameter tuning, and training of extreme learning machines (ELMs). In terms of hyperparameter optimization (HPO), the count of neural cells in the intermediate coat of the ELM network, accompanied by the initialization of weights and biases, will be taken into account. The proposed methods have been developed for improving IoT security and were therefore tested on a Windows 10 security from the TONIot base. bookPart Springer, Singapore 507 522 10.1007/978-981-99-1726-6_39 LNNS Lecture Notes in Networks and Systems: ICoISS 2023: International Conference on Intelligent Sustainable Systems, volume 665