Evaluation of the efficiency of reference machine learning models for buffer memory prediction when transforming a self-similar input stream of packets into a stream having exponential distribution under the condition of equality of mathematical expectations and median flows

Using machine learning methods, models have been developed to predict the size of the queue depending on the Hurst exponent based on the data obtained when performing the transformation of an input self-similar stream distributed according to the Pareto law into a stream having an exponential distribution with equal mathematical expectation and equal medians. A comparative analysis of the obtained models is carried out. Each model was examined using the following quality metrics: coefficient of determination, rms regression error, mean absolute error, penalty value, estimated loss. Models that use isotonic regression and support vector methods are the best in terms of the selected quality metrics for methods of transforming the input and output packet streams when the mathematical expectation is equal. For methods of transforming the input and output packet stream with equal medians, linear models are the best.

1. M. L. Fedorova, T. M. Ledeneva., Ob issledovanii svoistva samopodobiya trafika mul'tiservisnoi seti //Vestnik voronezhskogo gosudarstvennogo universiteta. Seriya: sistemnyi analiz i informatsionnye tekhnologii. 2010. №1. S.46-54.

2. Shelukhin O. I., Fraktal'nye protsessy v telekommunikatsiyakh / O.I. Shelukhin, A.M. Tenyakshev, A.V. Osin; Pod red. O.I. Shelukhina. - M.: Radiotekhnika, 2003 - 479 s.

3. Linets G.I., Govorova S.V., Voronkin R.A, Mochalov V.P., Imitatsionnaya model' asinkhronnogo preobrazovaniya samopodobnogo trafika v uzlakh kommutatsii s ispol'zovaniem ocheredi // Infokommunikatsionnye tekhnologii. 2019. T.17. №3. S. 293-303.

4. Linets G.I., Govorova S.V., Voronkin R.A. Funktsional'nye preobrazovaniya samopodobnogo potoka paketov s sokhraneniem znacheniya mediany. Sovremennaya nauka i innovatsii, 2021, №1. g. Pyatigorsk, C. 50-57.

5. Gennadiy Linets, Roman Voronkin, Svetlana Govorova, Ilya Palkanov, Carlos Grilo. The Regression Analysis of the Data to Determine the Buffer Size. YRID-2020: International Workshop on Data Mining and Knowledge Engineering. CEUR-WS. org, ISSN 1613-0073, Vol- 2842 – 150 pp.

6. Linets G.I., Govorova S.V., Voronkin R.A. Programma formirovaniya nabora dannykh dlya issledovaniya statisticheskikh kharakteristik modeli preobrazovaniya samopodobnogo trafika. Svidetel'stvo o gos. registratsii programmy dlya EHVM № 2019619275. Data registr. 15.07.19.

7. Handbook of Mathematics. Sixth Edition / I.N. Bronshtein, K.A. Semendyayev, G. Musiol, H. Mühlig.URL: https://doi.org/10.1007/978-3-662-46221-8

8. Bazovye printsipy mashinnogo obucheniya na primere lineinoi regressii. URL: https://habr.com/ru/company/ods/blog/322076/ (data obrashcheniya 01.06.2020).

9. Isotonic regression. URL: https://scikit-learn.org/stable/modules/isotonic.html (data obrashcheniya 01.06.2020).

10. Westling T., Gilbert P., Carone M. Causal isotonic regression. URL: http://arxiv.org/abs/1810.03269 (data obrashcheniya 23.05.2020).

11. Sharden B., Massaron L., Bosketti A. Krupnomasshtabnoe mashinnoe obuchenie vmeste s Python / per. s angl. A. V. Logunova. M.: DMK Press, 2018. 358 s.

12. Support Vector Regression (SVR) using linear and non-linear kernels. URL: https://scikit-learn.org/stable/auto_examples/svm/plot_svm_regression.html?highlight=svr (data obrashcheniya 01.06.2020).