الفهرس 
INTRODUCTIONOnly 14 pages are availabe for public view
 |  | from | 129 |  |  |
| | | from | 129 | | |
Abstract
The internet and cloud infrastructure have undergone a massive change in the last decade. Which lead to the need of Secure software for automated high-performance networks & clouds. Network traffic classification plays an increasingly key role in this rapidly changing environment. As a result, network services continuously offer new features, have non-stop increase in performance, and are continuously perfecting their resources. Which introduces not only new possibilities but also significant research challenges for future telecommunication networks and applications researchers. Software-based networks are a two-sided sword, with flexibility and agility on one side, and uncertain reliability/performance on the other side. it is an open challenge to ensure that these efforts will result into available networks/services, producing maximal performance with minimal resources in dynamically changing environments. Network traffic classification has made extensive use of machine learning methods. In this study, we used K-nearest neighbor, decision tree C4.5, and logistic regression. In order to optimize the performance of the overall classifier in the neural network mode, we have used artificial neural networks and convolutional neural networks, a process known as ensemble learning. This configuration produced accuracy as high as 92.55% and very little model loss in comparison to other study work; this would be a good addition to network classification efforts using machine learning.
The thesis is divided into five chapters including lists of contents, tables, and figures as well as a list of references.
Chapter 1
This chapter is an Introduction including the motivation for this work, followed by the thesis outline and contributions.
Chapter 2
This chapter includes a literature review on methods for traffic classification with focus on machine learning techniques and classified traffic types.
Chapter 3
This chapter proposes a new solution of ensemble-based network traffic classification. Simulation results are shown at the end of this chapter.
Chapter 4
This chapter presents deployments of network classifier, comparing old and contemporary traffic classification solutions and applications.
Chapter 5
This chapter Concludes the work of this thesis work. Suggested future work is presented.