Feature reduction is widely used in machine learning based phishing URL detection to lower model complexity while retaining detection performance. However, conclusions about how much reduction is acceptable may depend on how training and testing data are partitioned. This study examines whether conclusions about feature reduction remain stable when evaluation changes from random stratified to domain disjoint cross validation. Experiments were conducted on URL-Phish Version 2 using Mutual Information ranking within each training fold and four predefined feature budgets, from 22 to 5 features. Logistic Regression and Random Forest were evaluated under both protocols, with PR AUC as the primary metric. PR AUC decreased as the feature budget was reduced for both classifiers under both evaluation protocols, so the qualitative conclusion about feature reduction remained consistent within this experimental setting. However, the size of the protocol gap varied across classifiers and feature budgets: it was clearest for Random Forest under the smallest feature budget, whereas the Logistic Regression gaps were less clearly separated from fold-level variability. Domain disjoint evaluation also produced greater fold-level variability. The results show that claims about compact phishing URL representations should be interpreted together with the evaluation protocol used. Future work should examine whetherthis pattern persists across additional datasets, classifiers, and domain grouping rules.

Missing data are common in clinical prediction studies, yet their handling may affect not only predictive performance but also which model is judged best. This study examined the stability of classifier selection when missing data handling was changed under controlled, paired evaluation conditions. Using SUPPORT2 data from 9,105 patients, Logistic Regression, Random Forest, and XGBoost were evaluated with median, K nearest neighbor, and iterative imputation. The analysis retained natural missingness, added nested missing completely at random (MCAR) perturbations of 5%, 10%, 20%, and 30%, and included a separate 20% missing at random (MAR) condition. Repeated stratified five fold cross validation used the same patient partitions, model seeds, and artificial missingness masks across corresponding comparisons. Ranking stability was assessed through condition level winner changes, paired rank reversals, and agreement across receiver operating characteristic area under the curve (ROC AUC), Average Precision, and Brier Score. The condition level winner remained stable under natural and mild additional missingness, but became dependent on imputation at higher missingness. Winner changes occurred in 6 of 12 imputation comparisons and 9 of 15 missingness comparisons, while pairwise rank reversals occurred in about 44% of matched repeat comparisons. All three metrics selected the same winner in 8 of 18 conditions. Because competing winners were separated by small ROC AUC margins, the results indicate sensitivity of model selection rather than large performance advantages. Reporting ranking stability alongside conventional performance estimates may therefore provide a more cautious basis for comparative clinical prediction studies.

The current study explores the interconnectedness between financial development, information and communication technology (ICT), renewable energy and ecological sustainability in Tunisia over the 1991Q1–2021Q4 period. While there have been numerous studies that tested the determinants of ecological sustainability in Tunisia using indicators like carbon emissions (CO2) and ecological footprint (EF), a comprehensive evaluation of the load capacity curve (LCC) in context of Tunisia remains notably absent. This work fills the gap by examining the impact of financial development, ICT, and renewable energy on ecological sustainability within the LCC framework. The empirical findings reveal that the LCC hypothesis is valid in the context of Tunisia. In addition, ICT has been observed to have a positive effect on ecological sustainability by increasing load capacity factor (LCF). Furthermore, we found that natural resources and financial development have adverse effects on LCF. This study’s findings provide policymakers insights into the effects of financial development and ICT on LCF that is essential in formulating effective strategies and policies to achieve a sustainable environment.

This study proposes a new robust blind video watermarking technique for copyright protection that combines the Fourier transform, entropy analysis, Hessenberg decomposition, and lifting wavelet transform (LWT). The host video is divided into frames, the RGB channel of each frame is then isolated, and the 2-level LWT is applied to the R channel. The FFT coefficients are divided into 4x4 non-overlapping blocks after the FFT is applied to the high-frequency sub-band of the 2-level LWT. The watermark is then hidden in the first row and first column element of the upper Hessenberg matrix after each selected block has been subjected to Hessenberg decomposition. The proposed technique is extremely invisible and resilient to attacks, as demonstrated by numerous simulated experiments. Comparisons with similar scheme show that the proposed watermarking scheme outperforms the others in terms of invisibility and robustness

The rapid growth of Internet of Things (IoT) applications has increased the number of connected devices operating in shared wireless environments. IoT systems rely on sensors, actuators, embedded electronics, software, and internet connectivity to collect, exchange, and process data across sectors such as smart homes, healthcare, transportation, industrial systems, energy systems, and public infrastructure. However, the dense deployment of wireless IoT devices can create significant Electromagnetic Compatibility (EMC) challenges, especially when multiple transmitters and receivers operate in proximity or within the same or adjacent frequency bands. These conditions may lead to Electromagnetic Interference (EMI), which can degrade communication performance, reduce device reliability, and affect safety in critical environments. This study uses a regulatory review and conceptual framework development, drawing on literature synthesis, regulatory requirement extraction, analytical classification, regulatory mapping, and risk-based EMC categorisation. It identifies key EMC and EMI challenges in IoT applications, particularly in dense wireless and low-power communication environments. It analyses immunity requirements in critical sectors such as vehicular networks, healthcare, industry, and public infrastructure. It also develops a regulatory framework that defines stakeholder roles and classifies IoT connection methods, application areas, and device structures to support safer deployment and reduce interference risks.

Hereditary cancer syndromes (i.e. BRCA1/2-breast cancer) represent a small portion of around 6% of the population worldwide with genetic diseases, but their histopathological diagnosis is still difficult. In this paper, we present a clinical decision support system that combines an improved XGBoost classifier with SHapley Additive Explanations (SHAP) to provide high predictive accuracy and interpretable AI (XAI) in breast cancer diagnosis based on nuclear morphological features. The proposed approach employs a five-stage framework, namely: (1) Data collection from the Wisconsin Breast Cancer Diagnostic (WDBC) dataset, (2) missing value treatment by using multiple imputation by chained equations (MICE), (3) feature normalization, (4) biomarker selection using recursive feature elimination with cross validation (RFECV), and (5) class imbalance addressing by synthetic minority over-sampling technique (SMOTE). Hyperparameter tuning was done with Optuna using the Tree-structured Parzen Estimator (TPE) for 300 trials. The XGBoost model, evaluated on a test data consisting of 114 samples (20% of the total 569 samples), attained the highest accuracy of 96% and recall of 97% for class malignant, with a close to perfection ROC AUC of 0.99. SHAP results demonstrated that morphological characteristics (convex points and concavity, in particular) were the most important features, which is in line with histopathological understanding of nuclear alterations in hereditary cancer syndromes. Local waterfall plots ensured case-specific interpretability by breaking down single predictions into additive feature contributions. Although the WDBC dataset does not contain direct labels of genotype, the designed framework acts as a reproducible proof-of-concept for transparent, high-performance diagnosis of breast cancer with potential for application to hereditary syndromes, and veterinary histopathology incorporating appropriate species specific validation.