Probability distributions are essential statistical instruments, which make it possible to model and analyze some probability events in various spheres, including engineering, medicine, finance, and environmental science. Classical distributions, however, have been observed to have constraints in describing the complexity of real-life data. This paper will discuss these shortcomings by proposing the DUS-transformed Generalized Polynomial Quadratic Failure Rate (DUS-GPQF) distribution, which is a new extension of the generalized linear failure rate (GLF) distribution via DUS transformation method. The DUS-GPQF distribution increases the flexibility in the GLF model which provides it with greater ability to support a larger spectrum of data behaviors. The DUS-GPQF distribution has important statistical properties which include the hazard rate functional, moments, incomplete moments, entropy and the extropy. The DUS-GPQF distribution has seventeen estimation methods, which guarantee that it is practical to apply. The tests on the DUS-GPQF distribution are performed on two real-world data sets, and the results show that the model is better than other competitive models in goodness of fit and predictive accuracy. The study offers a powerful statistical model to a complex data so as to improve theoretical as well as practical statistical analysis.
The malicious domains and adversarially crafted URLs in cyber threats evolve at a very high speed. The detection frameworks need to be robust, adaptive, and scalable in such scenarios. Traditional detection mechanisms are static feature-based approaches that cannot perform well against unseen threats, adversarial manipulations, and long-term attack evolutions. Existing systems lack granular threat attribution, cross-organization intelligence sharing, and adversarial robustness, making them unsuitable for modern cyber defenses. To address these limitations, we introduce an LLM-Based Frequent Monitoring Framework that combines five advanced techniques: Meta-Learned Self-Supervised Domain Generalization (ML-SSDG), Reinforcement Learning-Augmented Adversarial Training (RL-AdvTrain), Hierarchical Multi-Task Threat Classification (HMT-TC), Temporal Memory-Augmented Transformer for Sequential Threat Detection (TMAT-STD), and Federated Privacy-Preserving Threat Intelligence Learning (FPPTIL). ML-SSDG can achieve zero-shot detection for novel attack domains with a reduction of false negatives by 20% and enhancement of zero-shot accuracy by up to 30%. RL-AdvTrain strengthens the model against masked malicious URLs, detecting 40% more adversarial threats. HMT-TC improves threat attribution and increases classification accuracy for attacks by 50%. TMAT-STD allows for identifying emerging domain threats in real-time, while such detection reduces the response time to domain-based malware campaigns by 30%. The last one is FPPTIL, which allows shared cross-organization threat intelligence without sharing the source of private data. The process of global threat detection improves by 30%. Our framework achieves a holistic, real-time, and privacy-preserving cyber defense solution that adequately outperforms traditional approaches in adversarial resilience, threat attribution, and zero-shot detections. Taken together, these improve the cybersecurity posture, reduce false positives, and support proactive mitigation of emerging cyber threats at scale in process.
In this paper, we consider the prediction problem of the future records based on observed data from two-parameter, shape and scale parameter, Kies distribution. Different point predictors including maximum likelihood, conditional median, best unbiased and Bayesian predictors of the future records are obtained. The corresponding prediction intervals using pivotal quantity, Highest Conditional Density (HCD), Shortest Length and Bayesian prediction intervals are also developed. The Monte Carlo algorithm is used to compute simulation consistent Bayesian prediction intervals for future unobserved records. The performance of the so obtained point predictors and prediction intervals are compared via experimental numerical simulation. The criteria that were considered for comparison purposes are mean square prediction error (MSPE) and prediction bias for point predictors and coverage probability (CP) and the average length (AL) for prediction intervals. A real and simulated data sets are performed for illustrative purposes.
Modern uncertainty-modeling frameworks-fuzzy sets, intuitionistic fuzzy sets, hyperfuzzy sets, neutrosophicsets, and plithogenic sets-provide powerful tools for capturing vagueness and imprecision. In particular, neutrosophic sets characterize each element by three independent degrees: truth, indeterminacy, and falsity. Classical neutrosophic sets have been refined by partitioning the membership degrees into additional components. Recently, the Hexapartitioned Neutrosophic Set, Octapartitioned Neutrosophic Set, Nonapartitioned Neutrosophic Set, and Decapartitioned Neutrosophic Set have been defined. In this paper, we introduce four novel partitioned models-hexapartitioned, octapartitioned, nonapartitioned, and decapartitioned neutrosophic offsets/oversets/undersets-and show how each can be seamlesslyembed ded into the plithogenic-set framework.
This paper presents a real-time web-based emotion recognition system based on unimodal deep learning models for facial and speech analysis, combined through decision-level score aggregation. Facial emotion recognition is performed using convolutional neural networks (CNNs), while speech emotion recognition relies on a CNN–BiLSTM architecture to capture both spatial and temporal speech patterns. These models are chosen for their effectiveness and low computational cost, making them suitable for web-based deployment. The facial model is trained on the FER2013 dataset, and the speech model is trained on the RAVDESS corpus using MFCC-based audio features. Rather than performing multimodal representation learning, this work demonstrates decision-level fusion by aggregating unimodal prediction scores to improve robustness when combining facial and speech information. Experimental results show competitive recognition performance and support the applicability of the proposed system for human-computer interaction in real-time and web-based affective applications.
Accurate flood forecasting remains a major challenge due to the nonlinear dynamics of hydrological processes and the difficulty of optimizing deep learning models. This study proposes a hybrid deep learning framework integrating Artificial Neural Networks (ANN), Long Short-Term Memory (LSTM), and Convolutional Neural Networks (CNN) with the Snake Optimization Algorithm (SOA) for hyperparameter tuning. The method includes feature normalization, training–testing partitioning, and multi-metric evaluation using MSE, RMSE, MAE, and R². The results reveal that the hybrid LSTM-SOA model achieved the best performance with R²= 0.8514, MSE=0.000386, RMSE=0.019653, and MAE=0.015849, outperforming standalone models. These results demonstrate the potential of hybrid optimisation-based deep learning as a trustworthy tool to support decisions in flood forecasting, early warning, and disaster preparedness.
This study introduces an innovative framework for addressing the fractional Fornberg–Whitham equation by melding the Yasser–Jassim integral transform with the Variational Iteration Method, all formulated under the Atangana–Baleanu fractional derivative in the Caputo interpretation. We first derive an explicit series representation of the solution and then rigorously prove that the iterative procedure converges, identifying conditions that guarantee both existence and uniqueness. In addition, we derive a bound on the truncation error to quantify the approximation’s accuracy. To validate the theoretical developments, a detailed computational example is provided, demonstrating rapid convergence and close agreement with the exact solution. The findings highlight the method’s robustness and suggest its broad applicability as an analytical tool for a wide range of nonlinear fractional partial differential equations.
Monkeypox (mpox) is a zoonotic infectious disease that has re-emerged as a global public health concern due to its increasing transmission in various regions. In this study, we propose a fractional-order epidemiological model to investigate the transmission dynamics of mpox involving human and rodent populations. The use of fractional-order derivatives allows the model to incorporate memory effects, which are relevant for capturing the long-term influence of past infections, immune responses, and exposure history. To evaluate effective intervention measures, an optimal control framework is developed by combining two time-dependent control strategies: human vaccination and rodent eradication. The optimal control problem is solved using Pontryagin's Principle of Minimum in conjunction with a forward-backward iterative algorithm, while the fractional-order system is numerically approximated using an Eulerian scheme. Model parameters are estimated using real mpox case data, and the performance of the fractional-order model is compared across different fractional-order values. Numerical simulations show that the combined control strategy significantly reduces the infected population and overall implementation costs compared to a single control intervention. Furthermore, the results show that higher fractional orders, approaching the integer order case, result in improved system performance and earlier separation between control strategies. These findings highlight the importance of memory effects in mpox transmission dynamics and provide insights for designing efficient and cost-effective intervention policies.
In the field of reliability and survival times, we note that many data are naturally limited above. The Exponentiated Mukherjee--Islam distribution (EMID) is considered one of the most important finite-range survival time distributions. It has two shape parameters and a scale parameter. Among its properties is that it can track the skewness and behavior of the hazard function while maintaining support at $(0, \theta)$. In this paper, more than one method was used to estimate both parameters and reliability of the EMID under type~II censoring. The model estimators were derived using maximum likelihood (ML) and maximum product spacing (MPS) methods. To demonstrate the efficiency of the estimators obtained in this paper, we presented an extended Monte Carlo simulation study in which each estimator was compared based on the value of the root mean square error (RMSE) using \texttt{R~Studio}. The simulation study used eight groups of parameters, sample sizes $n = 30, 50, 100, 300$, and censoring ratios $C = 0.0, 0.1, 0.2, 0.3$. The results show a decrease in RMSE with larger~$n$. High censoring ratios lead to an amplified RMSE, especially for the scaling parameter~$\theta$. Maximum likelihood estimates (MLE) are better in small samples with low censoring ratios, whereas MPS estimators tend to provide more robust and efficient results.
coronary artery disease (CAD) continues to be a major cause of death linked to cardiovascular issues, and thus early diagnosis is crucial to enhance patient outcome and prevent unnecessary medical interventions. Machine learning (ML) and data mining are increasingly being recognized as robust predictive methods for CAD, with opportunities for early detection and preventive medicine. This article discusses the role of various ML algorithms to predict CAD and enhance diagnostic performance, emphasizing the importance of such methodologies in medicine. The methodology includes a rigorous study of ML techniques such as neural networks, decision trees, support vector machines, and ensemble techniques like Random Forest and XGBoost. The paper explains the advantages and disadvantages of these techniques based on their applications with publicly available medical datasets to predict CAD. Data balancing algorithms such as SMOTE and ADASYN are also incorporated for improving model performance. The findings reveal that ensemble techniques, particularly XGBoost, register the highest accuracy (94.7%), closely trailed by Random Forest (92.04%). Additionally, data balancing techniques also enhance model recall and specificity to make predictions even more accurate. The findings point towards the power of sophisticated machine learning algorithms for CAD detection as well as the need for preprocessing data to reach maximum model efficiency. This research demonstrates the broader significance of machine learning for transforming CAD prediction, and the potential to improve patient care, reduce healthcare costs, and facilitate a shift toward preventative therapy in cardiovascular disease management also we will discuss artificial intelligence (AI) applications and Recent advances in AI with CAD.
The objective of this study was to examine the impact of Universal Health Coverage (UHC) on life expectancy, along with other determinants of the health status of the population in Morocco, using indicators representing health status measured by indicators such as education rates, urbanization rates, or indicators specific to access to healthcare and other variables divided into socioeconomic variables and healthcare system variables. As a method, we estimated an ARDL (AutoRegressive Distributed Lag) model on annual data representing the health status of the population in Morocco covering the period from 1990 to 2023. To do this, the ARDL (AutoRegressive Distributed Lag) approach was used to estimate the short-term and long-term dynamic relationship between life expectancy and its determinants using EViews software. As the results confirm the hypothesis of a long-term cointegration relationship and equilibrium between life expectancy and universal health coverage. They reveal that universal health coverage has a positive and statistically significant effect on life expectancy with a long-term coefficient of +1.249 (p<0.001). This estimate empirically validates the hypothesis that universal health coverage is a key determinant of improved population health.
Flooding is a recurrent problem in Banjar Regency due to its low-lying wetland topography, high rainfall, andriver sedimentation. Nonetheless, precise flood mapping continues to be a challenge due to the interference of cloud coveron optical imaging and spectral ambiguity in diverse wetland ecosystems. To address these limitations, this study proposesa hybrid SAR–optical method that combines Sentinel-1 backscatter with Sentinel-2 water indices (FWEI and AWEI) usingpixel-level feature stacking and feature contribution analysis. Four machine learning classifiers—Random Forest, LogisticRegression, Support Vector Machine (SVM), and XGBoost—were evaluated using a spatially independent validation schemeto ensure robust generalization. The results indicate that AWEI surpasses individual features in water discrimination, whereasthe combination of SAR and AWEI consistently enhances classification stability and spatial coherence. Among the assessedmodels, SVM and RF exhibit similar performance, with SVM attaining the optimal classification balance (OA = 0.98, Kappa= 0.97, F1 = 0.98) for water detection, while the difference lacks statistical significance. Flood inundation maps were derivedthrough change detection between independently classified pre- and post-flood water maps. The results indicate that multisensor integration enhances the delineation of flood-affected areas, particularly in complex wetland environments. However,the study also highlights that the reliability of flood inundation mapping is inherently dependent on the accuracy of pre- andpost-flood water classification as well as the temporal consistency of the input imagery. Furthermore, feature contributionanalysis combined with spatial validation reveals that AWEI provided the strongest standalone classification performance,while SAR contributed complementary structural information that improved classification robustness. These findingsdemonstrate that spatial validation effectively mitigates autocorrelation bias, leading to more reliable and generalizablewater classification performance for flood inundation analysis.
This study introduces a numerical approach for solving the Generalized Absolute Value Matrix Equation. The motivation of this work lies in the fact that such equations arise in various applied mathematical and engineering problems, where the presence of the absolute value term makes the system strongly nonlinear and difficult to solve using standard linear algebra techniques. The elementwise absolute value introduces a nonlinearity, which makes standard linear solvers unsuitable. A tailored two-step fixed-point iteration is developed and tested on problems of varying sizes. Numerical experiments demonstrate that, with a suitable relaxation parameter, the method achieves reliable convergence, maintaining robustness and accuracy even for moderately sized problems.
Suppose the set $W=\{s_1, s_2,\dots, s_k \}$ is a subset of the vertex set $V(G)$. The representation of a vertex $v$ of $G$ with respect to $W$ as follows \[r_m(v|W)=\{d(v,s_1), d(v,s_2),\dots, d(v,s_k)\}\] where $d(v,s_i)$ is the distance between the vertex $v$ with the vertices of set $W$ together with their multiplicities. The set $W$ is called the {\it m-resolving set} of $G$ if every vertices of $G$ have distinct representation with respect to $W$. If $G$ has an m-resolving set, then an m-resolving set having minimum cardinality is called a multiset basis and its cardinality is called the multiset dimension of $G$, denoted by $md(G)$. We say that $G$ has an infinite multiset dimension and we write $md(G)=\infty$. In this paper, we determine the multiset dimension of kayak paddles graph and cycles with chord.
This study investigates the evolution of bancassurance in Egypt, specifically examining the impact of regulatory frameworks on the performance and technical efficiency of the Misr Insurance Holding Company (MIHC). The research analyzes two pivotal eras: the "experimental" phase (2004-2013), marked by regulatory instability, and the "reactivation" phase (2014-2020), following the strategic 2013 decree by the Central Bank of Egypt (CBE) and the Financial Regulatory Authority (FRA). The methodology employs a three-tiered quantitative approach. First, descriptive statistics and Independent Samples T-tests reveal a dramatic surge in performance during the reactivation phase: average Return on Investment (ROI) rose from 3.5% to 19.04%, and Distributable Profit to Equity increased from 3.1% to 29.43%. Second, an Interrupted Time Series Analysis (ITSA) was utilized to assess the causal impact of the 2013 intervention, identifying a positive immediate level shift in performance indicators (β2 = 0.1744 for ROI). Third, a non-parametric Data Envelopment Analysis (DEA) under Variable Returns to Scale (VRS) was applied to evaluate technical efficiency. Empirical results from the DEA model indicate a profound structural transformation; while the experimental phase exhibited high volatility and technical slack, with efficiency scores as low as 0.080, the post-2013 era achieved a stabilized and superior efficiency profile, reaching the "efficiency frontier" (score of 1.000) in the majority of the observed years. Furthermore, administrative expenses were successfully optimized, dropping from a peak of 326% of premiums in 2004 to a stable average of 14.13% after the reactivation. The study concludes that regulatory stability is the primary driver of operational maturity and resource optimization in the Egyptian life insurance sector, providing a roadmap for future bank-insurance integration.
Multicollinearity among predictor variables remains a major challenge in regression analysis. This issue arises when predictors are highly correlated, leading to inflated variances of ordinary least squares (OLS) estimators and unstable coefficient estimates. Several remedial methods have been proposed to mitigate Multicollinearity, including ridge regression, the Liu estimator, and principal component regression. A critical factor determining the performance of shrinkage estimators such as the Liu estimator is the selection of an appropriate shrinkage parameter, denoted by $d$. This study proposes a novel method for estimating the optimal value of $d$. The performance of the proposed estimator was evaluated through Monte Carlo simulations under varying levels of {{Multicollinearity}} severity and sample size. The method was also applied to a real-world dataset. Results demonstrate that the proposed estimator achieves a substantially lower mean squared error ($MSE$) and mean absolute error ($MAE$) compared to existing estimators, indicating superior estimation accuracy and stability.
This paper studies a boundary value (BV) problem characterized by variable order(VO) Caputo fractionalderivatives to study the existence, uniqueness, and stability of solutions under well-defined boundary conditions. Fixedpoint theory is used, where the Banach contraction principle ensures the uniqueness of solutions, while the Krasnoselskiitheorem confirms their existence. Furthermore, the notion of Ulam-Hyers stability is used to investigate the response of thesolutions to small perturbations. Numerical examples are presented to illustrate the theoretical results and to validate theapproach under practical conditions. Additionally, an application concerning the evolution of features in three - dimensionalnoise fields is included. The highlights its engineering relevance, particularly in image processing and signal analysis, wheremodeling noise behavior and memory effects is important for tasks such as denoising and feature extraction
Agricultural extension services face significant challenges delivering effective training to heterogeneous farming populations with diverse knowledge levels and resource constraints. Traditional uniform training approaches result in inefficiencies where experienced farmers encounter redundant content while novice farmers struggle with excessive complexity. This research develops an adaptive learning system for agricultural extension using Painting Training-Based Optimization (PTBO), a human-inspired metaheuristic algorithm. A multi-objective optimization framework was formulated incorporating knowledge gain maximization, time efficiency, sequence validity, difficulty appropriateness, and knowledge coverage, subject to time, budget, and prerequisite and essential knowledge constraints. A quasi-experimental study with 75 wheat farmers in Irbid Governorate, Jordan (2024-2025), randomly assigned participants to PTBO-personalized (n=25), GA-personalized (n=25), and traditional-uniform (n=25) groups. PTBO demonstrated superior performance: 15.3% improvement in knowledge gain over GA (32.4 vs. 28.1 points), 29.9% faster convergence (87.3 vs. 124.6 iterations), 96.2% knowledge retention at four-week follow-up, and 80.0% practical adoption versus 69.7% for traditional methods. Novice farmers achieved normalized learning gains of 0.76 compared to 0.68 (GA) and 0.66 (traditional). The research provides a deployable framework demonstrating that metaheuristic optimization effectively addresses agricultural knowledge dissemination challenges while maintaining computational efficiency for resource-constrained contexts.
This study investigates the drivers of inflation in Nigeria, focusing on the influence of foreign currencies, oil prices, domestic economic policies and supply chain disruption. With Nigeria experiencing an unprecedented inflation rate, the study employs five ensemble learning algorithms; Decision Tree, Random Forest, AdaBoost, Bagging, and Gradient Boosting regressions to uncover the relationships between inflation rates and its potential drivers. These drivers considered are exchange rates of foreign currencies, oil prices, domestic economic policies (the removal of oil subsidies and the floating of the Naira), and supply chain disruptions linked to the ongoing Russian-Ukrainian War. To better capture the effect of oil subsidy removal, floating of the Naira and Russia–Ukraine war, the continuous time-based variables which created a time counters that track the number of periods since these events occurred was used. The Exchange Rate Pass-Through (ERPT) Theory and Cost-Push Inflation Theory provided the theoretical framework for this study. Data from the Central Bank of Nigeria, covering January 2012 to September 2024, were used and the time series cross validation was used to ensure robustness against temporal dependencies while multicollinearity analysis was carried out using the Variance Inflation Factor (VIF) followed by PCA for dimensionality reduction. To estimate the possible effect of outliers the result was presented with and without robust scaler. Preprocessing was carried out, hyperparameters tuning were performed using Grid Search algorithm, and model performance was evaluated using metrics such as Mean Absolute Error, R-squared, Root Mean Square Error, Huber loss and adjusted R-squared using five-fold cross-validation. The results reveal a significant positive relationship between foreign currency exchange rates and inflation \( p < .05 \) with evidence of multicollinearity among features and hence the Principal Component Analysis (PCA) was used in dimensionality reduction. The use of robust scaler was found to improve the performance of the Machine Learning algorithms and the Gradient Boost algorithms outperformed other machine learning algorithms with the least RMSE (1.2990), MAE (0.8132) and Huber loss (0.5036) and the highest values of \( R^2 \) (0.9671) and adjusted \( R^2 \)(0.9634). Additional analysis using feature importance, Shapley Additive exPlanations (SHAP) and Partial Dependence (PP) plots showed that domestic policies variables particularly the removal of fuel subsidy and the floating of the Naira policy had the most significant positive impact on inflation in Nigeria while exchange rate of CFA and other global currencies such as USD and Euro were found to have moderate impact on inflation. These findings underscore an urgent need to reinvest savings from the subsidy removal into developmental projects such as building of refineries while also providing incentives and financial support to cushion the effect of subsidy removal. The need to establish currency swap agreements with key trade partners particularly China will help stabilize the Naira and its exposure to foreign currency volatility. These strategies could help in improving external balances, support industrial growth, and promote long –term resilience of Nigerian economy.
The importance of using self-regression models is conditional after smoothing the variance with the fluctuations of the daily closing price of gold globally for the period 1/1/2023 until 26/12/2024, including the GARCH-M(p,q) and TEGARCH(p, q)models, and diagnosing the models with the problem of heterogeneity of variation, estimating the parameters of the models used in the greatest possible way, examining the models using tests and statistical criteria to obtain the best models that represent real data, and then processing them using the Daubechies Wavelet and the Symlets Wavelet, and examining the suitability of forecasting models, it turned out that processing data with a wave gives better results than in real data, since a model with fewer parameters was obtained, which is the TGARCH model(1,1).
