Development and evaluation of the effectiveness of an algorithm for automatic classification of network events
Keywords:
network event classification, machine learning, algorithm development, performance evaluation, NSL-KDD dataset, UNSW-NB15 dataset, random forestAbstract
With the increasing volume of network traffic and security threats, automatic classification of network events has become vital. This paper presents the development and evaluation of a machine learning-based algorithm for network event classification. The algorithm extracts statistical and payload-based features from network packets and applies feature selection techniques. Supervised learning models such as decision trees, random forest and neural networks are trained on the filtered feature sets. The algorithm is evaluated on NSL-KDD and UNSW-NB15 datasets using metrics like accuracy, precision and recall. Experimental results show that the random forest classifier achieves the best performance with over 95% accuracy on both datasets. The proposed algorithm demonstrates high effectiveness in classifying network events into benign and attack categories in real-time.
References
IsmagilovaE. et al. Smart cities: Advances in research—An information systems perspective. Int. J. Inf. Manage.(2019)
AguileraU. et al..Citizen-centric data services for smarter cities. Future Gener. Comput. Syst. (2017)
ZhangY. et al. HotML: A DSM-based machine learning system for social networks. J. Comput. Sci. (2018)
SapountziA. et al. Social networking data analysis tools & challenges. Future Gener. Comput. Syst.. (2018)
GuptaB.B. et al. Recent research in computational intelligence paradigms into security and privacy for online social networks (OSNs). Future Gener. Comput. Syst. (2018)
CostaK.A. et al. Internet of Things: A survey on machine learning-based intrusion detection approaches. Comput. Netw. (2019)
