Ajloun National University

Android malware detection systems face critical challenges including data scarcity for emerging threat families, high-dimensional feature spaces, and concept drift caused by evolving attack techniques. Traditional machine learning approaches require extensive labeled datasets and frequent retraining, limiting their practical deployment against rapidly emerging threats. This paper proposes an adaptive few-shot malware classification framework that integrates CatBoost-based feature selection, prototypical networks with episodic meta-learning, quantum-enhanced classification, concept drift detection, and explainable AI (XAI) analysis using SHAP and LIME. The CatBoost feature selection reduces dimensionality by 99.46% on CCCS-CIC-AndMal-2020 (9,503 to 51 features) and 94.07% on KronoDroid (489 to 29 features) while preserving discriminative information. The prototypical network learns metric-based representations enabling classification with only 5 support samples per class. Extensive experiments demonstrate state-of-the-art performance with 99.70% accuracy on CCCS-CIC-AndMal-2020 (15 malware families) and 99.33% accuracy on KronoDroid (binary classification), outperforming existing methods by 0.70-9.70%. The framework exhibits robust temporal stability with maximum accuracy degradation of 0.24% across evaluation periods. XAI analysis reveals that file descriptor manipulation and file system operations are the most discriminative features for malware detection. These results establish few-shot prototypical learning with intelligent feature selection as an effective paradigm for practical malware detection requiring minimal annotation, interpretable decisions, and stable long-term performance.

This study examined the fragmented legal framework governing compulsory insurance for autonomous vehicles in the United States and assessed its implications for liability allocation and compensation for victims. Guided by three core research questions—(1) what is the doctrinal and statutory basis for mandating compulsory insurance on autonomous vehicle manufacturers, (2) how does state-level regulatory variation affect liability allocation and compensation mechanisms for cross-jurisdictional incidents, and (3) which federal legislative approaches are legally viable to harmonize insurance and liability rules—the analysis demonstrated that the absence of a harmonized federal approach created significant legal uncertainty and inefficiencies. The research analyzed judicial precedents, statutory developments, and prior legislative initiatives at both the federal and state levels. Building on these findings, the paper proposes a federal legislative model that aims to unify liability rules, establish minimum compulsory coverage, mandate the disclosure of accident data, and create a federally administered compensation fund. These proposals aimed to enhance legal clarity, ensure adequate protection for victims, and facilitate the safe integration of autonomous vehicles into the national transportation system.

The exponential growth of Internet of Things (IoT) deployments demands fog computing architectures that position computational resources at the network edge to ensure low-latency processing. However, optimal fog node placement in dynamic environments remains a significant challenge due to node mobility, random failures, and fluctuating traffic patterns, which require continuous adaptation to maintain service quality. This paper introduces the Dynamic Pufferfish Optimization Algorithm (D-POA), a bio-inspired metaheuristic that leverages behavioral strategies observed in pufferfish species to adaptively reposition fog nodes in real-time, balancing exploration of the solution space with refinement of promising configurations. D-POA formulates the placement problem as a continuous multi-objective optimization that simultaneously addresses network connectivity, area coverage, and movement costs to minimize service disruption during reconfiguration. Comprehensive experimental evaluation across five dynamic scenarios demonstrates that D-POA achieves 97.8% connectivity and 98.4% coverage while reducing movement costs by 38-57% compared to baseline algorithms, with near-linear scalability maintaining over 96% solution quality across networks ranging from 50 to 1000 nodes.

The Internet of Things (IoT) faces critical security challenges stemming from resource-constrained devices and inadequate intrusion detection capabilities. Traditional machine learning approaches struggle with high-dimensional network traffic data due to the curse of dimensionality, severe class imbalance between benign and malicious traffic, and dependence on manual feature engineering that fails to capture complex non-linear attack patterns. Although deep neural networks offer automatic feature extraction, they suffer from two fundamental limitations: the degradation problem, where increasing network depth paradoxically raises training error rather than improving performance, and uniform channel weighting, which prevents the network from adaptively emphasizing attack-relevant features while suppressing irrelevant noise. This research proposes a novel hybrid framework integrating Fractional Tchebichef moment-based feature preprocessing with deep Residual Networks enhanced by Squeeze-and-Excitation (ResNet-SE) attention mechanisms. Fractional Tchebichef moments provide compact, noise-resistant representations by operating directly in the discrete domain, eliminating discretization errors inherent in continuous moment approaches. Network traffic features are transformed into 232 × 232 moment-based matrices capturing discriminative patterns across multiple scales. Comprehensive evaluation on Bot-IoT and Leopard Mobile IoT datasets demonstrates superior performance, achieving 99.78% accuracy and a 99.37% F1-score, substantially outperforming K-Nearest Neighbors (84.7%), Support Vector Machines (87.5%), and baseline CNNs (99.3%). Ablation studies confirm synergistic contributions, with residual connections contributing 0.18% and SE attention adding 0.14% improvements. Cross-dataset evaluation achieves 96.34% and 97.12% accuracy on UNSW-NB15 and IoT-Bot datasets without retraining, while the framework processes 127.9 samples per second across diverse attack taxonomies.

Polymorphism is a cornerstone of object-oriented programming that promotes code reuse and interface consistency. While runtime polymorphism through virtual functions is well understood, compile-time polymorphism via function overloading deserves deeper study in the context of modern C++. This paper provides a detailed analysis of function overloading as a mechanism for compile-time polymorphism. It explores overload-resolution rules, evaluates performance differences between static and dynamic dispatch, and integrates these findings with features introduced up to C++20. The benchmarking experiments are grounded in a computational geometry case study that compares dispatch mechanisms through area calculations for geometric shapes. Experimental results show that static dispatch through overloading outperforms virtual function calls by a factor of 3.2 in micro-benchmarks. The study concludes that combining overloading with modern C++ features enables high-performance, type-safe, and expressive code.

This study aims to verify the role of artificial intelligence in developing engineering project management. It seeks to determine the feasibility of implementing artificial intelligence to increase productivity, improve safety, reduce costs, and save time within engineering projects. The focus of the study is on two approaches to project management: traditional, which does not involve extensive integration of the latest intelligent systems, and innovative, where artificial intelligence is used for automated decision-making and risk forecasting. The methodological tools include pairwise comparison methods based on the Saati scale, as well as formulas for integral assessment, which compare the total benefits and opportunities with costs and risks. The results indicate that the innovative approach with deep integration of artificial intelligence has a higher overall indicator due to better productivity, more efficient resource allocation, and a more flexible security system, despite the additional initial costs and risks of implementation. In the long term, this approach allows for significant time savings and improved economic performance, which is critically important in the context of global competition and rapid technological change. This study confirms the feasibility of using artificial intelligence and provides an analytical tool for rational decision-making between traditional and innovative approaches to engineering project management.

Abstract The current study examines the inclined magnetic flow of nanoparticles with radiative flow and the Cattaneo-Christov dual diffusion model. The analysis involves the mixture of alumina Al 2 O 3 nanoparticles with (Carboxymethyl cellulose-water/ Al 2 O 3 ). In addition, this study examines how the arrangement and size of alumina nanoparticles affect the behaviour of nanofluid flow. The thermal Biot, solutal Biot, thermal source, and heat space dependent are also incorporated. The Cattaneo-Christov heat and mass flux formulas are used to make a gradual change to the heat and concentration equations. The current issue is represented using partial differential equations (PDEs), which are then transformed into ordinary differential equations by appropriate similarity transformations. The acquired highly nonlinear system of modeled equations has been solved by consuming the numerical approach through the boundary value problem Bvp4c method. The primary objective of this research is to examine the characteristics of velocity, temperature, and concentration in a nanofluid, specifically focusing on the impact of the size and spacing of aluminum nanoparticles. In addition, the calculation of heat and mass transfer is performed for the Cattaneo-Cristove dual diffusion. Velocity distribution has been depreciated for the enhancing values of magnetic field and porosity for interspacing and radius of nanoparticles. Moreover, drag friction is lower due to the larger values of the magnetic field and porosity variable.

Background: Ujjayi pranayama and osteopathic manipulative treatment (OMT) both enhance respiratory function and musculoskeletal alignment. For individuals with Upper Cross Syndrome these methods can have a significant impact. Objective: to investigate how much does adding Ujjayi Pranayama into osteopathic manipulation (OM) on pulmonary functions and functional capacity in college students diagnosed with upper cross syndrome (UCS). Methods: Two-armed, parallel-group RCT with concealed allocation and blinded assessment of outcomes conducted on forty participants with UCS randomly divided into two groups. Ujjayi pranayama and OM administered to the experimental group while the control group received only osteopathic manipulation. Both groups was undergo 45-minute sessions, five times a week for eight weeks. We measured ERV, PEF, MVV and IC by portable spirometer. Results: Adding ujjayi pranayama to OM showed significant greater improvements in PEF (MD change was 0.76±0.07 L/s vs. 0.49±0.09 L/s in control (p=0.025)), ERV (MD change 0.29±0.06 L vs. 0.08±0.01 L in control (p=0.002)), 6-Minute Walk Distance Test (6 MWDT) ( MD change23.60±2.05 meters vs. 14.70±1.54 meters in control (p=0.001), TUG (MD change was -2.20±0.20 sec vs. -1.15±0.11 sec in control (p=0.001). Some parameters showed no statistically significant differences in changes between groups: MVV: Difference and % change was not significant (p=0.250 and p=0.461). IC: Borderline difference in mean change (p=0.058) but % change was not significant (p=0.201). Conclusion: Ujjayi pranayama to OM resulted in superior improvements in both respiratory functions in upper cross syndrome, rather than osteopathy alone.

This study investigates the influence of Environmental Regulations (ER) and Financial Incentives (FI) on Supply Chain Efficiency (SCE) and Green Logistics Implementation (GLI) among small and medium-sized enterprises (SMEs) in Jordan. Drawing on the theoretical foundations of Sustainable Supply Chain Management (SSCM), this study proposes and empirically tests a conceptual model linking ER and FI to SCE and GLI. A quantitative research design was employed using data collected from 243 professionals across Jordanian SMEs involved in logistics and supply chain operations. The data were analyzed using Partial Least Squares Structural Equation Modeling (PLS-SEM) via SmartPLS 4. The results reveal that both ER and FI significantly and positively influence SCE and GLI, though with varying magnitudes. ER exerted a stronger effect on SCE (β = 0.372, p < 0.001) compared to GLI (β = 0.172, p < 0.01), suggesting that regulatory compliance enhances operational efficiency but may not fully drive proactive green logistics practices. Conversely, FI demonstrated the strongest impact on SCE (β = 0.412, p < 0.001) and a weaker yet significant effect on GLI (β = 0.148, p < 0.05), indicating that financial support serves as a critical enabler of efficiency but requires complementary institutional and cultural mechanisms to promote sustainability adoption. These findings underscore the complementary roles of regulatory and financial instruments in advancing both efficiency and sustainability within SME supply chains. The results also provide practical insights for policymakers and SME managers aiming to strengthen the competitiveness and sustainability of Jordan’s supply chain ecosystem.

This research examined how prepared Jordanian university students are for e-learning with artificial intelligence (AI) tools, concentrating on the impacts of digital literacy, previous experience, attitudes, and infrastructure. A quantitative survey methodology was utilized, with a student population of 400 individuals across various academic disciplines. All measurement scales demonstrated a high degree of internal consistency (α = .84–.90) as determined using a reliability analysis. High readiness levels were revealed through descriptive results, with attitudes being the most dominant and infrastructure being the most challenging. All variables retained positive correlation, and multiple regression analysis revealed that the leading indicators of readiness were attitudes and digital literacy, followed by previous experience and infrastructure. Gender was not a significant readiness variable, but ANOVA results showed variation depending on students’ academic disciplines. Results indicate that the primary factors influencing readiness for AI-enabled e-learning are non-technical and psychological, as opposed to demographic factors. Suggested actions include the provision of digital literacy training, infrastructure development, and discipline-oriented AI educational resources to ensure equitable integration of AI in higher education, and more proactive measures to build a positive attitude towards AI.