USING NEURAL NETWORKS IN NETWORK SECURITY PREDICTION
Keywords:
SVM algorithm, K-Means algorithm, Apriori algorithm, fuzzy clustering algorithm, neural networkAbstract
The article analyzes four algorithms: SVM, fuzzy clustering, K-Means and Apriori. We have described in detail the four stages of ensuring the security of network users and controlling their access. Research on a specially created reliable model for predicting network security.Intrusion pattern detection based on neural networks was developed, capable of identifying anomalies and attacks associated with abuse. This model performs three types of classification tasks: distinguishing between an attack and a normal state, as well as between ideal types of attacks or a normal state. In addition, the model demonstrates the classification accuracy, execution speed and amount of memory used. The main models include achieving high accuracy, reducing processing time and minimizing memory consumption. The proposed model based on neural networks successfully meets these goals. In the modern world, networks are becoming increasingly complex, closely interconnected and widely applicable. Network traffic volumes are growing almost exponentially, making networks more vulnerable to attacks by attackers who want to disrupt their functioning. Such vulnerabilities threaten economic losses and the leakage of confidential information. Therefore, there is an urgent need to improve methods for detecting vulnerabilities and improving the quality of network security prediction. A network security prediction model has been developed, aimed at reducing memory consumption and increasing the speed and accuracy of detecting various types of attacks. The results showed that the model is characterized by low memory consumption, fast attack detection time, and high accuracy. The methods used to create this model are characterized by simple implementation. They are also cost-effective, since the use of neural networks does not require additional costs. The model simplifies calculations, which makes it an effective solution for predicting network security.Thus, neural networks are a recommended tool for developing such models. In the future, it is planned to improve the models for more accurate and faster intrusion detection.
References
Qiuhua, W., Xiaoqin, O., & Jiacheng, Z. (2019). A classification algorithm based on data clustering and data reduction for intrusion detection system over big data. KSII Transactions on Internet and Information Systems, 13(8), 3714–3732.
Jianliang, M., Haikun, S., & Ling, B. (2009). The application on intrusion detection based on K-means cluster algorithm. In 2009 International Forum on Information Technology and Applications (pp. 150–152). Chengdu, China.
Aung, Y. Y., & Myat, M. M. (2018). Hybrid intrusion detection system using K-Means and classification and regression trees algorithms. In 2018 IEEE 16th International Conference on Software Engineering Research, Management and Applications (SERA) (pp. 195–199). Kunming, China.
Ravale, U., Marathe, N., & Padiya, P. (2015). Feature selection based hybrid anomaly intrusion detection system using K-Means and RBF kernel function. Procedia Computer Science, 45, 428–435.
Xiaoyi, H. (2020). Network security situation prediction based on grey relational analysis and support vector machine algorithm. International Journal of Network Security, 22(2), 177–182.
Xiao, Y., Wang, H., & Xu, W. (2015). Parameter selection of Gaussian kernel for one-class SVM. IEEE Transactions on Cybernetics, 45(5), 927–939.
Song, C., & Ma, K. (2009). Design of intrusion detection system based on data mining algorithm. In 2009 International Conference on Signal Processing Systems (pp. 370–373). Singapore.
Rochim, A. F., Aziz, M. A., & Fauzi, A. (2019). Design log management system of computer network devices infrastructures based on ELK stack. In 2019 International Conference on Electrical Engineering and Computer Science (ICECOS) (pp. 338–342). Batam Island, Indonesia.
Jin, C., Ye, Z., Wang, C., Yan, L., & Wang, R. (2018). A network intrusion detection method based on hybrid rice optimization algorithm improved fuzzy C-means. In 2018 IEEE 4th International Symposium on Wireless Systems within the International Conferences on Intelligent Data Acquisition and Advanced Computing Systems (IDAACS-SWS) (pp. 47–52). Lviv, Ukraine.
Han, J. W., & Kamber, M. P. (2011). Data mining: Concepts and techniques (3rd ed.). Elsevier Science.
Huang, Z. (2019). Research and implementation of intrusion detection based on host log. In Proceedings of the International Conference on Big Data Engineering (pp. 98–106).
Lei, X., Chunxiao, J., Jian, W., Jian, Y., & Yong, R. (2014). Information security in big data: Privacy and data mining. IEEE Access, 2, 1149–1176.
Pratik, M., Saunil, D., & Ravina, D. (2015). Intelligent security systems – Intrusion detection system. Retrieved from https://github.com/jgera/Network-Intrusion-Detection-System/blob/master/report.pdf
Цвик О. С. (2023). Аналіз і особливості програмного забезпечення для контролю трафіку. Вісник Хмельницького національного університету. Cерія: Технічні науки, (1).
Твердохліб А. О., Коротін Д. С. (2022). Ефективність функціонування комп’ютерних систем при використанні технології блокчейн і баз данних. Таврійський науковий вісник. Серія: Технічні науки, (6).
