Search In this Thesis
   Search In this Thesis  
العنوان
Development machine learning techniques to enhance cyber security algorithms /
المؤلف
Amer, Ghada Mohamed.
هيئة الاعداد
باحث / غادة محمد عامر
مشرف / محمد عبدالعظيم محمد
مشرف / إيهاب هاني عبدالحي
مشرف / إبراهيم ياسر عبدالباسط
مناقش / أحمد شعبان مازن سمرة
الموضوع
Machine learning. Artificial intelligence. Computer security.
تاريخ النشر
2022.
عدد الصفحات
online resource (99 pages) :
اللغة
الإنجليزية
الدرجة
ماجستير
التخصص
الهندسة الكهربائية والالكترونية
تاريخ الإجازة
1/1/2022
مكان الإجازة
جامعة المنصورة - كلية الهندسة - قسم الاتصالات والالكترونيات
الفهرس
Only 14 pages are availabe for public view

from 99

from 99

Abstract

”Nowadays, cyber security plays an important role in the field of information technology (IT), where cyber security threats are a growing global problem. With the development of technology, electronic threats are increasing, as securing information has become one of the biggest challenges facing the information society, especially with the recent developments in the fields of cloud computing.Distributed Denial of Service (DDoS) is one of the most dangerous attacks faced in cloud computing. This attack aims to make cloud services unavailable to end-users by depleting system resources, resulting in huge losses that pose a threat to national security and information security assets, and thus making the development of defensive solutions against such attacks necessary to expand the use of cloud computing technology.Using machine learning (ML) is one way to secure cloud computing. Machine learning (ML) has shown promising results in detecting cyber-attacks including DDoS when applied to cloud intrusion detection systems.In this study, the proposed system is built using Random Forest (RF) as a supervised machine learning algorithm, which is a collective learning method that works by building many decision trees at training time. Experiments using the most common and standard datasets, NSL-KDD and CICIDS2017, achieved detection accuracy of 99.69% for the first dataset and 99.97% for the second dataset, respectively. The proposed system performs well when compared to other methods in terms of accuracy, detection rate, and low false-positive rate.”