Abstract
A new memory structure for artificial neural networks of adaptive-resonance theory is proposed, which has a hierarchical form. For each new memory level, the previous classification value is refined by increasing the similarity parameter. This architecture was used to determine the network state in intrusion detection systems. The paper describes an algorithm for learning the proposed structure of the ART-2m network in parallel mode. A comparative analysis of the time characteristics of the network with the proposed structure when operating in series and parallel modes is carried out. Experiments were carried out using the NSL KDD-2009 sample, the results of which show the possibility of using ART-2m in intrusion detection systems. Most states were determined with sufficient accuracy that is greater than 80%.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Positive Technologies. Vulnerabilities of corporate information systems, p. 21 (2018)
Petrenko, S.A.: Methods for detecting intrusions and anomalies in the functioning of cyber systems. Proc. Inst. Syst. Anal. Russian Acad. Sci. 41, 194–202 (2009)
Duque, S., bin Omar, M.N.I.: Using data mining algorithms for developing a model for intrusion detection system (IDS). Procedia Comput. Sci. 61, 46–51 (2015)
Ghosh, S., et al.: A novel neuro-fuzzy classification technique for data mining. Egypt. Inform. J. 15(3), 129–147 (2014)
Idhammad, M., Afdel, K., Belouch, M.: Distributed intrusion detection system for cloud environments based on data mining techniques. Procedia Comput. Sci. 127, 35–41 (2018)
Emelyanova, U.G., et al.: Neural network technology for detecting network attacks on information resources. Softw. Syst.: Theor. Appl. 2(3) (2011)
Markov, R.A., et al.: Research of neural network technologies to identify information security incidents. Young Sci. 23, 55–60 (2015)
Carpenter, G.A., Grossberg, S.: ART 2: self-organization of stable category recognition codes for analog input patterns. Appl. Opt. 26(23), 4919–4930 (1987)
Carpenter, G.A., Grossberg, S., Rosen, D.B.: ART 2: an adaptive resonance algorithm for rapid category learning and recognition. Neural Netw. 4(4), 493–504 (1991)
Bukhanov, D.G., Polyakov, V.M.: Adaptive resonance theory network with multilevel memory. Sci. Bull. Belgorod State Univ. 45(4), 665–673 (2018)
The NSL‐KDD Data Set. [Electronic resource]. http://www.unb.ca/cic/datasets/nsl.html. Accessed 20 Jan 2021
Aggarwal, P., Sharma, S.K.: Analysis of KDD dataset attributes-class wise for intrusion detection. Procedia Comput. Sci. 57, 842–851 (2015)
Kajemskiy, M.A., Shelukhin, O.I.: Multi-class classification of network attacks on information resources using machine learning methods. Proc. Educ. Inst. Commun. 5(1), 107–115 (2019)
Acknowledgment
This work was supported by Russian Foundation for Basic Research, project 19-29-09056 mk and Ministry of Education of Russia (grant IS) project No. 13.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Polyakov, V., Bukhanov, D., Panchenko, M. (2021). Identification of the Network State Based on the ART-2 Neural Network with a Hierarchical Memory Structure in Parallel Mode. In: Kovalev, S.M., Kuznetsov, S.O., Panov, A.I. (eds) Artificial Intelligence. RCAI 2021. Lecture Notes in Computer Science(), vol 12948. Springer, Cham. https://doi.org/10.1007/978-3-030-86855-0_3
Download citation
DOI: https://doi.org/10.1007/978-3-030-86855-0_3
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-86854-3
Online ISBN: 978-3-030-86855-0
eBook Packages: Computer ScienceComputer Science (R0)