Smart User Consumption Analysis Using Support Vector Machine and Multilayer Perceptron


Published: 2023-07-12

Page: 163-170

Madugu Jeremiah Omanga

Department of Computer Science, Taraba State University, Jalingo, Nigeria.

Olajide Blessing Olajide *

Department of Computer Engineering, Federal University, Wukari, Nigeria.

Okpor James

Department of Computer Engineering, Federal University, Wukari, Nigeria.

Andrew Ishaku Wreford

Department of Computer Science, Federal University, Wukari, Nigeria.

*Author to whom correspondence should be addressed.


The advent of smart devices and the rapid growth of wireless and wired networks has brought about a swift change in the methods of entertainment and information sources. This rapid expansion is due to the rise in over-the-top (OTT) usage, among which are the use of social networks, gaming, online video, and audio applications. OTT delivers audio, video, and other media content over the internet generating high traffic on the network infrastructure. Degradation of service is a typical tactic used to address excessive consumption patterns of smart users. However, it is deficient in that, rarely is the analysis of consumer behavior taken into consideration. Analysis of user’s consumption habits before the degradation of service is necessary for the efficiency of this mechanism. This study proposes a smart user consumption analysis using Support Vector Machine (SVM) and Multilayer Perceptron (MLP). Real-world traffic data collected in the network from Universidad Del Cauca, Popayán with a total of 2,704,839 flow instances was used. SVM and MLP were used in this study to analyze and classify user consumption behavior on OTT applications since user consumption behavior in real-world networks changes over time. The experimental result of this study showed that the two models performed extremely well in the analysis and classification of OTT application users. Nevertheless, MLP out performed SVM in both precision and F1-score. The evaluation metrics used are Recall, Precision, and F1-score based on True Positive (TP), False Positive (FP), True Negative (TN), and False Negative (FN).

Keywords: Smart devices, OTT application, network, degradation, traffic, SVM, MLP

How to Cite

Omanga , M. J., Olajide , O. B., James , O., & Wreford , A. I. (2023). Smart User Consumption Analysis Using Support Vector Machine and Multilayer Perceptron. Asian Research Journal of Current Science, 5(1), 163–170. Retrieved from


Download data is not yet available.


Gupta, G., and Singharia, K. Consumption of OTT media streaming in COVID-19 lockdown: Insights from PLS analysis. Vision. 2021;(25)1:36-46.

Korotky S., “Semi-Empirical Description and Projections of Internet Traffic Trends Using a Hyperbolic Compound Annual Growth Rate,” Bell Labs Tech. J. 2013;18(3):5–21.

Karimiyazdi R, Mokhber M. Improving Viral Marketing Campaign via Mobile Instant Messaging (MIM) Applications. Proc. 1st World Virtual Conf. Social and Behavioural Sciences. 2015;1–13.

Sudtasan T, Mitomo H. Effects of OTT services on consumer’s willingness to pay for optical fiber broadband connection in Thailand. presented at the 27th Eur. Regional Conf. Int. Telecommun. Soc. (ITS), Evol. North-South Telecommun. Divide, Role Europe, Cambridge, U.K.; 2016 [Online].


Jain MK. The Rise of OTT Platform: Changing Consumer Preferences. EPRA International Journal of Multidisciplinary Research (IJMR). 2021;(7)6:257-261.

Rojas JS, Pekar A, Rendón Á, Corrales JC. Smart user consumption profiling: Incremental learning-based OTT service degradation. IEEE access. 2020;8:207426-207442.

Chetty M, Banks R, Brush AJ, Donner J, Grinter R. You're capped: understanding the effects of bandwidth caps on broadband use in the home. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems. 2012;3021-3030.

Chetty M, Kim H, Sundaresan S, Burnett S, Feamster N, Edwards WK. ucap: An internet data management tool for the home. In Proceedings of the 33rd annual ACM conference on human factors in computing systems. 2015;3093-3102.

Olajide BO, Wreford AI. “A Predictive Model for Car-Loan Repayment Credibility of Customers”, Asian Basic and Applied Research Journal. 2023;1(5):81-89.

Radha R. Characterisation of youtube video streaming traffic, a masters thesis submitted to Department of Electrical Engineering, School of Computing Bliekinge Institute of Technology, Kariskrona, Sweden. 2013;1- 50.

Vinupaul MV, Bhattacharjee R, Rajesh R, Kumar GS. User characterization through network flow analysis. In 2016 International Conference on Data Science and Engineering (ICDSE). 2016;1-6.

Rojas JS, Gallón ÁR, Corrales JC. Personalized service degradation policies on OTT applications based on the consumption behavior of users. In International Conference on Computational Science and Its Applications. 2018;543-557.

Bakhshandeh A, Eskandari Z. An efficient user identification approach based on Netflow analysis. In 2018 15th International ISC (Iranian Society of Cryptology) Conference on Information Security and Cryptology (ISCISC) 2018;1-5.

Rojas JS, Rendon A, Corrales JC. Consumption behavior analysis of over the top services: Incremental learning or traditional methods? IEEE Access, 7, 2019; 136581-136591.

Choi, J., and Kim, Y. A heterogeneous learning framework for over-the-top consumer analysis reflecting the actual market environment. Applied Sciences. 2021;(11)11:4783.