Members of the Department D. Peraković, M. Periša and I. Cvitić in collaboration with fellow researcher Brij B. Gupta (parent institution: Department of Computer Engineering, National Institute of Technology Kurukshetra, Kurukshetra, Haryana, India) published a scientific paper entitled "Ensemble machine learning approach for classification of IoT devices in smart home" in the renowned scientific journal International Journal of Machine Learning and Cybernetics. The scientific paper was published in the Open Access and is available free of charge via the link.
The impact of this scientific journal is visible through its indexing (Web of Science Core Collection, Web of Science Current Content) and the impact factor for 2019 = 3,753, and according to Scimago belongs to the first quartile (Q1).
The published paper is the result of research on IoT devices' behavior in a smart home environment where we were able to define device classes based on traffic characteristics generated by IoT devices and we developed a model for classifying such devices into previously defined classes. The classification model was developed using the logitboost method of supervised machine learning and shows a high classification efficiency> 99% according to all validation measures characteristic of classification models.
The obtained results open space for further research in this interesting area. What is important to us as a research team is that it provides us with a strong foothold for the development of new cyber-attack detection systems specific to the IoT paradigm.
