Designing an Offline-First Digital Aid Distribution System for Post-Conflict Contexts: A Case Study of Tigray, Ethiopia
E-Governance System: A Bibliometric Exploration for Current Trend, Gaps, and Future Directions
Implementation of Intrusion Detection System with UNSW-NB15 Dataset using Variants of CNNs
Intelligent Video Learning Assistant using LLaMA Model
Online Shopping: A Context-Aware, Advanced Machine Learning Algorithms and Architecture
Humanitarian aid delivery in post-conflict settings often suffers from systemic inefficiencies such as reliance on paper- based processes, fragmented data management, and degraded infrastructure. This study proposes a tailored offline- first Digital Aid Distribution System (DADS) for the Tigray region of Ethiopia, where widespread infrastructural damage complicates aid logistics. Using a mixed-methods approach—including stakeholder consultations, literature review, and Agile prototyping—the research defines core requirements for a system capable of functioning amid intermittent connectivity and limited resources. The proposed architecture employs a three-tier model featuring a mobile client for field operations, a centralized server for data management, and a synchronization mechanism for intermittent connectivity. A functional prototype demonstrates modules for beneficiary management, inventory tracking, distribution execution, and analytics. Simulations using the prototype suggest potential reductions in distribution cycle time (approximately 45%) and reporting time (around 96%), alongside improvements in transparency and beneficiary satisfaction. The study concludes that a context-aware digital system is not only feasible but essential for enhancing the efficiency, accountability, and dignity of aid delivery in Tigray. Recommendations include forming strategic partnerships, adopting phased piloting, and exploring integration with national digital identity systems and financial technologies.
E-governance has emerged as a critical mechanism for transforming public administration through the effective use of digital technologies, enhancing transparency, efficiency, and citizen participation in governance processes. Despite the rapid growth of scholarly output in this domain, a comprehensive understanding of its intellectual structure, research trends, and emerging themes remains limited. This study presents a bibliometric analysis of e-governance research published between 2004 and 2025, drawing on data retrieved from the Scopus database. Following a rigorous screening process based on predefined inclusion criteria, 63 peer-reviewed journal articles were selected for analysis. Bibliometric and science mapping techniques, including citation analysis, keyword co-occurrence, co-authorship, and collaboration network analysis, were employed using VOSviewer and Biblioshiny software. The findings reveal a steady growth in e-governance research, particularly after 2015, reflecting the increasing integration of information and communication technologies in public administration. The Netherlands, India, the United States, and the United Kingdom emerged as key contributors in terms of research productivity and citation impact. Core thematic clusters highlight e-governance, e-government, governance approaches, ICT, public administration, and citizen participation as dominant research areas. However, the analysis also identifies conceptual gaps related to emerging technologies, ethical concerns, digital inequality, and sustainability. This study provides a systematic overview of the evolution of e- governance research, offering valuable insights for scholars and policymakers while outlining promising directions for future research.
The rapid evolution of network traffic and cyber-attack sophistication has necessitated robust Intrusion Detection Systems (IDS). Traditional machine learning methods often struggle to capture complex non-linear attack patterns. Deep learning, particularly Convolutional Neural Networks (CNNs), provides an effective alternative for automatic feature extraction and accurate classification. This paper presents the implementation of an IDS using the UNSW-NB15 dataset and explores various CNN architectures — including 1D-CNN, 2D-CNN, and hybrid CNN-LSTM models. A detailed experimental comparison is carried out in terms of accuracy, precision, recall, F1-score, false positive rate (FPR), and false negative rate (FNR). The results show that the hybrid CNN-LSTM model outperforms conventional CNN variants, achieving an accuracy of 98.6% for binary classification and 96.1% for multi-class detection. The study demonstrates the potential of CNN-based architectures to efficiently detect modern network intrusions.
This paper presents an intelligent video learning assistant that leverages advances in artificial intelligence and natural language processing to enhance learning from video-based educational content. The proposed system automatically extracts transcripts from online videos, generates concise summaries, and creates contextually relevant quiz questions using the LLaMA-3 large language model. By integrating transcript summarization, automated question generation, and semantic answer evaluation, the system transforms passive video consumption into an interactive learning experience. The architecture employs FastAPI for backend processing, React.js for a responsive user interface, and Fireworks AI for efficient model inference. Experimental observations demonstrate that the system reduces the time required to extract key concepts from long videos while improving learner engagement through immediate feedback and self-assessment. The proposed solution is particularly beneficial for students, educators, and professionals seeking an efficient and structured approach to video-based learning.
Online shopping websites play a vital role in modern digital commerce by enabling users to browse, compare, and purchase products conveniently. With the rapid growth of product catalogs, user data, and transactional information, traditional rule-based systems struggle to provide personalized experiences, efficient search, and intelligent recommendations. Modern e-commerce platforms leverage advanced technologies such as artificial intelligence, data analytics, recommendation systems, and secure backend architectures to enhance user engagement, search accuracy, and overall shopping experience. This paper presents a comprehensive analysis of an online shopping website, focusing on system architecture, core functionalities, and performance optimization techniques. The study includes product management, user authentication, shopping cart functionality, order processing, and secure payment integration. AI-driven features such as personalized product recommendations, smart search, sentiment analysis of reviews, and automated notifications significantly improve customer satisfaction and business efficiency.
