Romanian Journal of Information Science and Technology (ROMJIST)

An open – access publication

  |  HOME  |   GENERAL INFORMATION  |   ROMJIST ON-LINE  |  KEY INFORMATION FOR AUTHORS  |   COMMITTEES  |  

ROMJIST is a publication of Romanian Academy,
Section for Information Science and Technology

Editor – in – Chief:
Radu-Emil Precup

Honorary Co-Editors-in-Chief:
Horia-Nicolai Teodorescu
Gheorghe Stefan

Secretariate (office):
Adriana Apostol
Adress for correspondence: romjist@nano-link.net (after 1st of January, 2019)

Founding Editor-in-Chief
(until 10th of February, 2021):
Dan Dascalu

Editing of the printed version: Mihaela Marian (Publishing House of the Romanian Academy, Bucharest)

Technical editor
of the on-line version:
Lucian Milea (University POLITEHNICA of Bucharest)

Sponsor:
• National Institute for R & D
in Microtechnologies
(IMT Bucharest), www.imt.ro

ROMJIST Volume 27, No. 3-4, 2024, pp. 336-347, DOI: 10.59277/ROMJIST.2024.3-4.07
 

Khirota YALDA, Diyar Jamal HAMAD, Nicolae TAPUS, and Ibrahim Taner OKUMUS
Network Traffic Prediction Performance Using LSTM

ABSTRACT: As networks expand to support various applications involving text, audio, video, and images, data traffic increases correspondingly. Traffic classification, which identifies the origin of observed traffic, has multiple applications, including dynamic bandwidth allocation, traffic analysis, quality of service, and network security. Traditional network traffic classification methods like deep packet inspection rely on manually creating and maintaining communication profiles for various applications. However, these methods face challenges such as dynamic port changes and encrypted traffic. Machine Learning (ML) classifiers offer effective solutions to these issues, providing accurate network traffic classification. Due to these advancements, deep learning models are now utilized for network traffic classification and prediction. Long Short-Term Memory (LSTM) has emerged as a highly effective deep learning technique for addressing time series prediction challenges. This study aims to analyze the performance of forecasting network traffic using LSTM, with different activation functions and optimizers with Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE) and Coefficient of Determination (R-Squared) parameters as the model evaluation index. This demonstrates how these parameters impact network traffic forecasting performance.

KEYWORDS: Activation function; LSTM; optimization algorithms; traffic prediction

Read full text (pdf)






  |  HOME  |   GENERAL INFORMATION  |   ROMJIST ON-LINE  |  KEY INFORMATION FOR AUTHORS  |   COMMITTEES  |