Information and Computer Science‎

Permanent URI for this collection

https://hdl.handle.net/20.500.11888/10130

Browse

Recent Submissions

Now showing 1 - 5 of 175
  • No Thumbnail Available
    Item
    AI-Powered Network Intrusion Detection via Packet Sniffing
    (2025-09-04) Ahmad Sawafta
    As networks become increasingly complex cyberattacks such as port scanning, distributed denial-of-service (DDoS) attacks, and injection-based vulnerabilities, persistently challenge the CIA of data. Traditional signature-based Intrusion Detection Systems (IDS) are often inadequate against novel attack vectors, generating high false positives and struggling with scalability. This paper presents the design, implementation, and comparative evaluation of a machine learning-based IDS that performs real-time traffic analysis through packet sniffing. The system captures network packets, preprocesses them to extract salient features, and employs a dynamic, multi-model analysis engine. We rigorously evaluated several machine learning algorithms, including Logistic Regression, Support Vector Machines (SVM), on a mixture of normal and malicious traffic datasets. Logistic Regression and SVM offered an optimal balance of high accuracy, low computational overhead, and minimal false positives, making them exceptionally suitable for real-time deployment. The proposed system demonstrates its scalability and adaptability for modern network security. Keywords: Network Security, Intrusion Detection System (IDS), Machine Learning, Packet Sniffing, Real-time Analysis, Logistic Regression, SVM
  • No Thumbnail Available
    Item
    Real Time Ransomware Detection and Mitigation Using Machine Learning
    (2025-06-18) Bahaa Nofal
    Abstract Ransomware remains one of the most dangerous cybersecurity threats, causing severe damage by encrypting user data and demanding payment. Traditional antivirus systems often fail to detect new or evolving ransomware strains due to their reliance on signature-based detection. In this project, we present an intelligent, real-time ransomware detection and mitigation system that leverages behavioral analysis and machine learning. The system monitors file activities, registry operations, and network behavior using C++ and Windows APIs, and extracts technical features from executable files. These features are then classified using a trained Random Forest model in Python to determine whether a file is benign or malicious. If ransomware is detected, the system takes immediate action to stop the threat before it causes harm. The integration between C++ and Python enables efficient real-time classification, while the system’s modular design ensures extensibility and adaptability to new threats. Our solution was tested on real-world samples in a controlled virtual environment, and achieved high accuracy in detecting ransomware before execution.
  • No Thumbnail Available
    Item
    0xZone: A Secure Cybersecurity Challenge Platform
    (2025-06-18) Ali Abdulnaser Tawfiq Al-Abwah
    This thesis presents the design, implementation, and security evaluation of 0xZone, a web-based platform for hosting Capture The Flag (CTF) cybersecurity competitions in educational environments. The platform addresses security challenges in educational technology while providing a foundation for practical cybersecurity training and assessment. The research focuses on web application security, secure system design principles, and deployment strategies applied to educational technology platforms. The primary challenges addressed include: authentication and authorization in multi-user environments, protection against common web application vulnerabilities, basic anti-cheat mechanisms for competitive integrity, and secure containerized deployment in educational network infrastructures. The platform implements a multi-layered security architecture using modern web development practices, secure coding standards, and basic security monitoring. Key features include: containerized deployment with Docker, role-based access control, secure session management, input validation and sanitization, audit logging capabilities, and automated scanner detection for common security tools. Security evaluation demonstrates the platform’s resilience against common web application vulnerabilities including injection attacks, cross-site scripting, and session hijacking. The platform provides a functional foundation for educational cybersecurity training with appropriate security controls for academic environments. The research contributes to the field of educational technology by demonstrating practical implementation of security controls in web applications, providing a framework for secure CTF platform deployment, and establishing practices for protecting educational competition data.
  • No Thumbnail Available
    Item
    Deer Balak Alena
    (2025-06-18) Farah Assaf; Lial Awwad
    Deer Balak Alena which means "Take care of us" in the Palestinian dialect) is a public safety system designed to enhance security and emergency response in Palestine. The platform delivers real-time risk alerts and combats misinformation by verifying the credibility of news reports using AI-powered mechanisms. The system enables users to report incidents such as roadblocks, security threats, and emergency situations. Verified alerts are then distributed to both residents and emergency services through multiple communication channels, including offline methods, ensuring accessibility even without an internet connection. The system aims to create a secure and reliable digital space for emergency communication by implementing advanced trust and security mechanisms. These ensure the authenticity of users and content, verify submitted reports, and uphold community standards. By integrating predictive analytics, the system can anticipate potential risks and assist both authorities and individuals in making informed decisions. Strong security protocols are applied to protect user data and maintain system integrity. Through a combination of alert notifications, emergency coordination, and misinformation control, Deer Balak Alena plays a vital role in building a safer and more informed community.
  • No Thumbnail Available
    Item
    Mzad Palestine
    (2025-06-19) Waleed Dweikat; Karim Mithqal; Saad Odeh
    Abstract MzadPalestine is a machine learning-powered auction platform designed to modernize traditional auction practices for Palestinian communities. Integrated with payment gateways like Strip, the platform serves two primary stakeholders: buyers (bidders) and sellers, alongside administrators. The system streamlines auction workflows through real-time bidding, ML-driven price predictions, and secure payment processing. Key features include a dark mode interface, post-auction review moderation, and compliance with GDPR/PCI-DSS standards for secure transactions. Designed as a responsive web application, MzadPalestine prioritizes accessibility, transparency, and cultural relevance, with plans for mobile app development in future phases. The platform supports dual transaction models: real-time auctions and fixed-price "Buy Now" purchases. Expanding beyond traditional auctions, MzadPalestine integrates comprehensive jobs and services marketplaces. The jobs module connects employers with job seekers, supporting both traditional employment sectors and emerging digital opportunities. The services marketplace enables local professionals and craftspeople to offer their expertise, from technical services to cultural crafts, with secure booking systems. These additional modules leverage the same robust infrastructure, and community-focused approach as the core auction platform. By bridging technology with localized practices, the platform aims to reduce information asymmetry, enhance trust, and drive economic empowerment across Palestinian regions.