Abstract
Many modern machine learning tools are inefficient due to the pronounced nonlinearity of traffic changes and nonstationarity. For this, the task of predicting the signs of increments (directions of change) of the process of time series is singled out. This article proposes the use of some results of the theory of random processes for a quick assessment of the predictability of signs of increments with acceptable accuracy. The proposed procedure is a simple heuristic rule for predicting the increment of two neighboring values for a random sequence. The connection of this approach to time series with known approaches to the prediction of binary sequences is shown. The possibility of using the experience of predicting the absolute values of traffic in predicting the signs of changes is considered.
REFERENCES
Tong, H., Li, C., and He, J., A boosting-based framework for self-similar and non-linear internet traffic prediction, Advances in Neural Networks-ISNN 2004, Yin, F.L., Wang, J., and Guo, C., Eds., Lecture Notes in Computer Science, vol. 3174, Berlin: Springer, 2004, pp. 931–936. https://doi.org/10.1007/978-3-540-28648-6_148
Chen, A., Law, J., and Aibin, M., A survey on traffic prediction techniques using artificial intelligence for communication networks, Telecom, 2021, vol. 2, no. 4, pp. 518–535. https://doi.org/10.3390/telecom2040029
Dalmazo, B.L., Vilela, J.P., and Curado, M., Performance analysis of network traffic predictors, J. Network Syst. Manage., 2017, vol. 25, pp. 290–320. https://doi.org/10.1007/s10922-016-9392-x
Shimall, T., Traffic analysis for network security: Two approaches for going beyond network flow data, 2016. https://citeseerx.ist.psu.edu/viewdoc/download?doi= 10.1.1.513.7546&rep=rep1&type=pdf.
Zhao, A. and Liu, Yi., Application of nonlinear combination prediction model for network traffic, Proc. 2nd Int. Conf. on Electronic and Mechanical Engineering and Information Technology, Atlantis, 2012, pp. 2337–2340. https://doi.org/10.2991/emeit.2012.519
Lysyak, A.S. and Ryabko, B., Time series prediction based on data compression methods, Probl. Inf. Transm., 2016, vol. 52, no. 1, pp. 92–99. https://doi.org/10.1134/s0032946016010075
Sornette, D. and Andersen, J., Increments of uncorrelated time series can be predicted with a universal 75% probability of success, Int. J. Mod. Phys. C, 2000, vol. 11, no. 4, pp. 713–720. https://doi.org/10.1142/s0129183100000626
Frenkel, S., Theoretical aspects of a priori on-line assessment of data predictability in applied tasks, Cyber Security Cryptography and Machine Learning. CSCML 2021, Dolev, S., Margalit, O., Pinkas, B., and Schwarzmann, A., Eds., Lecture Notes in Computer Science, vol. 12716, Cham: Springer, 2021, pp. 187–195. https://doi.org/10.1007/978-3-030-78086-9_14
Viola, R., Martin, A., Morgade, J., Masneri, S., Zorrilla, M., Angueira, P., and Montalban, J., Predictive CDN selection for video delivery based on LSTM network performance forecasts and cost-effective trade-offs, IEEE Trans. Broadcasting, 2021, vol. 67, no. 1, pp. 145–158. https://doi.org/10.1109/tbc.2020.3031724
Ikharo, A.B., Anyachebelu, K.T., Blamah, N.V., and Abanihi, V.K., Optimising self-similarity network traffic for better performance, Int. J. Sci. Res. Sci. Technol., 2020, vol. 7, no. 4, pp. 164–176.
Feder, M., Merhav, N., and Gutman, M., Universal prediction of individual sequences, IEEE Trans. Inf. Theory, 1992, vol. 38, no. 4, pp. 1258–1270. https://doi.org/10.1109/18.144706
Ryabko, B. and Monarev, V., Using information theory approach to randomness testing, J. Stat. Planning Inference, 2005, vol. 133, no. 1, pp. 95–110. https://doi.org/10.1016/j.jspi.2004.02.010
Bosq, D. and Nguyen, H.T., A course in stochastic processes, A Course in Stochastic Processes: Stochastic Models and Statistical Inference, Theory and Decision Library B, vol. 34, Dordrecht: Springer, 1996. https://doi.org/10.1007/978-94-015-8769-3
Box, G.E.P., Jenkins, G.M., and Reinsel, G.C., Time Series Analysis: Forecasting and Control, New York: John Wiley & Sons, 2008.
Volovich, K.I., Deniso, S.A., Shabanov, A.P., and Malkovsky, S.I., Aspects of the assessment of the quality of loading hybrid high-performance computing cluster, CEUR Workshop Proc., 2019, pp. 7–11.
Christoffersen, P. and Diebold, F., Financial asset returns, direction-of-change forecasting, and volatility dynamics, Manage. Sci., 2022, vol. 52, no. 8, pp. 1273–1287. https://doi.org/10.1287/mnsc.1060.0520
Campbell, J.Y., Lo, A.W., and MacKinlay, A.C., The Econometrics of Financial Markets, Princeton Univ. Press, 1997.
Sharma, S., Activation functions in neural networks, 2019. https://towardsdatascience.com/activation-functions-neural-networks-1cbd9f8d91d6.
Pan, C., Wang, Yu., Shi, H., Shi, J., and Cai, R., Network traffic prediction incorporating prior knowledge for an intelligent network, Sensors, 2022, vol. 22, no. 7, p. 2674. https://doi.org/10.3390/s22072674
Frenkel, S., Predicting the direction of changes in the values of time series for relatively small training samples, Cyber Security, Cryptology, and Machine Learning. CSCML 2022, Dolev, S., Katz, J., and Meisels, A., Eds., Lecture Notes in Computer Science, vol. 13301, Cham: Springer, 2022, pp. 118–134. https://doi.org/10.1007/978-3-031-07689-3_9
ACKNOWLEDGMENTS
We express our gratitude to Prof. V.Yu. Korolev for participating in the discussion of the proposed sign prediction heuristic.
Funding
This work was supported by ongoing institutional funding. No additional grants to carry out or direct this particular research were obtained.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
The authors declare that they have no conflicts of interest.
Additional information
Translated by L. Solovyova
Publisher’s Note.
Allerton Press remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Frenkel, S.L., Zakharov, V.N. Internet Traffic Prediction Model. Sci. Tech. Inf. Proc. 50, 397–405 (2023). https://doi.org/10.3103/S0147688223050052
Received:
Published:
Issue Date:
DOI: https://doi.org/10.3103/S0147688223050052