Claim Missing Document
Check
Articles

Found 9 Documents
Search

Revolutionizing Cybersecurity: The GPT-2 Enhanced Attack Detection and Defense (GEADD) Method for Zero-Day Threats Jones, Rebet; Omar, Marwan
International Journal of Informatics, Information System and Computer Engineering (INJIISCOM) Vol 5 No 2 (2024): INJIISCOM: VOLUME 5, ISSUE 2, DECEMBER 2024
Publisher : Universitas Komputer Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34010/injiiscom.v5i2.12741

Abstract

The escalating sophistication of cyber threats, particularly zero-day attacks, necessitates advanced detection methodologies in cybersecurity. This study introduces the GPT-2 Enhanced Attack Detection and Defense (GEADD) method, an innovative approach that integrates the GPT-2 model with metaheuristic optimization techniques for enhanced detection of zero-day threats. The GEADD method encompasses data preprocessing, Equilibrium Optimization (EO)-based feature selection, and Salp Swarm Algorithm-Based Optimization (SABO) for hyperparameter tuning, culminating in a robust framework capable of identifying and classifying zero-day attacks with high accuracy. Through a comprehensive evaluation using standard datasets, the GEADD method demonstrates superior performance in detecting zero-day threats compared to existing models, highlighting its potential as a significant contribution to the field of cybersecurity. This study not only presents a novel application of deep learning for cyber threat detection but also sets a foundation for future research in AI-driven cybersecurity solutions
The Human Factor in Cybersecurity: Addressing the Risks of Insider Threats Zangana, Hewa Majeed; Sallow, Zina Bibo; Omar, Marwan
Jurnal Ilmiah Computer Science Vol. 3 No. 2 (2025): Volume 3 Number 2 January 2025
Publisher : PT. SNN MEDIA TECH PRESS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58602/jics.v3i2.37

Abstract

In the rapidly evolving landscape of cybersecurity, the human element remains one of the most critical and complex factors to manage. Insider threats, whether originating from malicious intent or inadvertent actions, pose significant risks to organizational security. This paper explores the multifaceted nature of insider threats, examining the motivations and behaviors that drive individuals to compromise systems. By analyzing case studies and current research, we identify key vulnerabilities and the role of organizational culture in mitigating these risks. Furthermore, we propose comprehensive strategies for detecting, preventing, and responding to insider threats, emphasizing the importance of continuous education, robust access controls, and advanced monitoring technologies. This paper aims to provide a holistic understanding of the human factor in cybersecurity and offers practical solutions to address the pervasive challenge of insider threats.
Cloud Architectures for Distributed Serverless Computing: A Review of Event-Driven and Function-as-a-Service Paradigms Zangana, Hewa Majeed; Sallow, Zina Bibo; Omar, Marwan
International Journal of Artificial Intelligence & Robotics (IJAIR) Vol. 6 No. 2 (2024): November 2024
Publisher : Informatics Department-Universitas Dr. Soetomo

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25139/ijair.v6i2.8597

Abstract

The advent of serverless computing has revolutionized the cloud computing landscape, providing scalable, cost-effective, and flexible solutions for modern application development. This paper comprehensively reviews cloud architectures for distributed serverless computing, focusing on event-driven and Function-as-a-Service (FaaS) paradigms. This research explores the fundamental principles and benefits of serverless computing, highlighting its impact on development practices and infrastructure management. The review covers key components, including orchestration, scalability, and security, and examines leading serverless platforms and frameworks. Through critically analyzing current research and industry practices, identify challenges and propose future directions for optimizing serverless architectures. This paper aims to explain how event-driven and FaaS paradigms reshape cloud computing, enabling developers to build resilient and efficient applications without server management. Our research found that event-driven architectures in serverless computing offer significant advantages in scalability, real-time processing, and resource utilization. FaaS paradigms provide modularity, granularity, and cost-effectiveness, making them suitable for various applications. Cloud-edge collaborative architectures are crucial for achieving low-latency and high-performance serverless applications but require robust security, privacy, and resource management frameworks.
TextGuard: Identifying and Neutralizing Adversarial Threats in Textual Data Albtsoh, Luay; Omar, Marwan
International Journal of Informatics, Information System and Computer Engineering (INJIISCOM) Vol. 6 No. 2 (2025): INJIISCOM: VOLUME 6, ISSUE 2, DECEMBER 2025 (ONLINE FIRST)
Publisher : Universitas Komputer Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Adversarial attacks inside the text domain pose a serious risk to the integrity of Natural Language Processing (NLP) systems. In this study, we propose "Text-Guard," a unique approach to detect hostile instances in natural language processing, based on the Local Outlier Factor (LOF) algorithm. This paper compares TextGuard's performance against that of more traditional NLP classifiers such as LSTM, CNN, and transformer-based models, while also experimentally verifying its effectiveness on a variety of real-world datasets. TextGuard significantly surpasses earlier state-of-the-art methods like DISP and FGWS, with F1 recognition accuracy scores as high as 94.8%. This sets a new benchmark in the field as the first use of the LOF technique for adversarial example identification in the text domain
Unveiling the Potential of Local Outlier Factor in Credit Card Fraud Detection Jones, Angel; Omar, Marwan
International Journal of Informatics, Information System and Computer Engineering (INJIISCOM) Vol 7 No 1 (2026): INJIISCOM: VOLUME 7, ISSUE 1, JUNE 2026 (ONLINE FIRST)
Publisher : Universitas Komputer Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

This study evaluates the Local Outlier Factor (LOF) algorithm for credit card fraud detection, emphasizing its effectiveness with imbalanced datasets. Unlike traditional methods that struggle with the rarity and variability of fraudulent transactions, LOF uses local density deviations to identify anomalies. Through a rigorous methodology involving data preprocessing, parameter tuning, and comparison with other machine learning algorithms, LOF demonstrated a high recall rate and a balanced precision-recall trade-off, excelling at detecting subtle, localized fraud. Challenges like threshold setting and false positives were noted, with future research suggested on real-time system integration, algorithm combination, and advanced feature engineering. The study underscores LOF's strengths and limitations, contributing to enhanced fraud detection strategies
A Novel Hybrid Algorithm for Effective Image Restoration Zangana, Hewa; Firas, Mahmood Mustafa; Omar, Marwan
Vokasi Unesa Bulletin of Engineering, Technology and Applied Science Vol. 2 No. 2 (2025)
Publisher : Universitas Negeri Surabaya or The State University of Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26740/vubeta.v2i2.38118

Abstract

Image restoration plays a pivotal role in various applications, from medical imaging to satellite photography, by enhancing the quality of images degraded by noise, blur, or other distortions. Traditional methods and deep learning techniques have both shown promise in addressing these challenges, yet each has its limitations. Traditional algorithms often struggle with complex distortions, while deep learning models demand extensive computational resources and large datasets. To harness the strengths of both approaches, we propose a novel hybrid algorithm that integrates traditional image restoration techniques with advanced deep learning models. This paper presents a novel hybrid algorithm for image restoration, integrating traditional Wiener filtering with a state-of-the-art U-shaped transformer (Uformer) architecture. Unlike existing methods, our approach combines the computational efficiency of classical techniques with the robustness and precision of deep learning. Comprehensive evaluations on benchmark datasets demonstrate significant improvements in restoration quality (PSNR/SSIM) and computational efficiency compared to state-of-the-art methods. This research contributes a new perspective on hybrid methodologies, bridging the gap between traditional and modern approaches in image restoration.
Small Object Detection in Medical Imaging Using Enhanced CNN Architectures for Early Disease Screening Zangana, Hewa Majeed; Omar, Marwan; Li, Shuai; Al-Karaki, Jamal N.; Vitianingsih, Anik Vega
Buletin Ilmiah Sarjana Teknik Elektro Vol. 7 No. 3 (2025): September
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/biste.v7i3.14015

Abstract

Early detection of subtle pathological features in medical images is critical for improving patient outcomes but remains challenging due to low contrast, small lesion size, and limited annotated data. The research contribution is a hybrid attention-enhanced CNN specifically tailored for small object detection across mammography, CT, and retinal fundus images. Our method integrates a ResNet-50 backbone with a modified Feature Pyramid Network, dilated convolutions for contextual scale expansion, and combined channel–spatial attention modules to preserve and amplify fine-grained features. We evaluate the model on public benchmarks (DDSM, LUNA16, IDRiD) using standardized preprocessing, extensive augmentation, and cross-validated training. Results show consistent gains in detection and localization: ECNN achieves an F1-score of 88.2% (95% CI: 87.4–89.0), mAP@0.5 of 86.8%, IoU of 78.6%, and a low false positives per image (FPPI = 0.12) versus baseline detectors. Ablation studies confirm the individual contributions of dilated convolutions, attention modules, and multi-scale fusion.However, these gains involve higher computational costs (≈2× training time and increased memory footprint), and limited dataset diversity suggests caution regarding generalizability. In conclusion, the proposed ECNN advances small-object sensitivity for early disease screening while highlighting the need for broader clinical validation and interpretability tools before deployment.
Hybrid Decision Support Framework with Explainable AI and Multi-Criteria Optimization Zangana, Hewa Majeed; Hassan, Noor Salah; Omar, Marwan; Al-Karaki, Jamal N.
Sistem Pendukung Keputusan dengan Aplikasi Vol 4 No 2 (2025)
Publisher : Ali Institute or Research and Publication

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55537/spk.v4i2.1328

Abstract

Decision-making in domains such as healthcare, finance, and smart systems demands frameworks that combine model-driven expertise with data-driven adaptability. This paper proposes a hybrid decision support framework that integrates Explainable AI (XAI) with multi-criteria optimization to enhance transparency, robustness, and adaptability. Unlike traditional systems, our approach unifies mechanistic models with machine learning and embeds interpretability and optimization mechanisms. Comparative evaluation against state-of-the-art methods shows consistent performance gains, achieving 15–25% lower error rates compared with data-driven baselines and generating more diverse Pareto-optimal solutions. These improvements highlight the framework’s potential as a reliable, explainable, and scalable solution for complex, real-world decision-making
From Generative AI to Objective-Driven Systems: A Paradigm Shift in Artificial Intelligence Omar, Marwan
The Indonesian Journal of Computer Science Vol. 14 No. 5 (2025): The Indonesian Journal of Computer Science
Publisher : AI Society & STMIK Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33022/ijcs.v13i5.4381

Abstract

The rapid advancement of Artificial Intelligence (AI), particularly in the domain of generative models, has led to impressive achievements in content creation and natural language processing. However, these models are inherently limited by their reliance on pattern recognition and lack of true understanding. In contrast, Objective-Driven AI offers a promising alternative by focusing on goal-oriented behavior, causal reasoning, and the development of world models. This paper explores the limitations of generative AI, highlighting its inability to grasp context, causality, and ethical considerations. It then presents the concept of Objective-Driven AI, emphasizing its potential to operate effectively in complex, real-world environments where understanding and reasoning are critical. The paper concludes with a discussion of future research directions, including advanced world modeling techniques, ethical AI, and robustness against adversarial attacks, which are essential for the further development of Objective-Driven AI systems. Keywords Objective-Driven AI, Generative AI, Causal Reasoning, World Modeling, Ethical AI, Artificial Intelligence, Adversarial Attacks, Machine Learning, Autonomous Systems