Search Results (2,053)

CBRN incidents are characterized by high uncertainty in terms of agent identity, dissemination methods, and situational context. This unpredictability complicates effective and timely response, especially in the initial phase before specialist services arrive, and lays the burden of applying protection and response measures on members of civil society participating in the incident. This paper proposes a structured classification framework for CBRN hazards to address this gap, integrating key characteristics from existing systems such as the GHS (Globally Harmonized System), WHO (World Health Organization) biosafety levels, and radiological exposure guidelines. The system emphasizes properties relevant for first responders and non-specialists, including observable effects, exposure routes, and hazard endpoints such as toxicity, virulence, and radiation dose. The goal is to enable rapid hazard recognition, improve communication, and support situational decision-making in public security scenarios. Full article

Multi-variant execution (MVX) is an active defense technique that can detect unknown attacks by comparing the outputs of redundant program variants. Despite notable progress in MVX techniques in recent years, current approaches for recovery of abnormal variants still face fundamental challenges, including state inconsistency, low recovery efficiency, and service disruption of an MVX system. Therefore, a record–replay-based state recovery approach for variants in MVX systems is proposed in this paper. First, a Syscall Coordinator (SSC), composed of a recording module, a classification module, and a replay module, is designed to enable state recovery of variants. Then, a synchronization and voting algorithm is presented. When an anomaly is identified through voting, the abnormal variant is handed over to the SSC for state recovery, while the Synchronization Queue is updated accordingly. Furthermore, to ensure uninterrupted system service, we introduce a parallel grouped recovery mechanism, which enables the execution of normal variants and the recovery of abnormal variants to run in parallel. Experimental results on SPEC CPU 2006 benchmark and server applications show that the proposed approach achieves low overhead in both the recording and replay phases while maintaining high state recovery accuracy and supports uninterrupted system service. Full article

The COVID-19 pandemic reshaped travel patterns and customer expectations, generating profound challenges for the hotel industry. This study analyzes 50,000 TripAdvisor reviews of New York hotels to examine how customer satisfaction with hotel selection attributes shifted before and during the pandemic. BERTopic was applied to extract eight key attributes, while VADER, PRCA, and Asymmetric Impact–Performance Analysis (AIPA) were used to capture asymmetric effects and prioritize improvements. Comparative analyses by hotel classification, travel type, and customer residence reveal significant shifts in food and beverage, location, and staff, particularly among lower-tier hotels, business travelers, and international guests. The novelty of this study lies in integrating BERTopic and AIPA to overcome survey-based limitations and provide a robust, data-driven view of COVID-19’s impact on hotel satisfaction. Theoretically, it advances asymmetric satisfaction research by linking text-derived attributes with AIPA. Practically, it offers actionable guidance for hotel managers to strengthen hygiene, expand contactless services, and reallocate resources effectively in preparation for future crises. In addition, this study contributes to sustainability by showing how data-driven analysis can enhance service resilience and support the long-term socio-economic viability of the hotel industry under global crises. Full article

Background and Objectives: The COVID-19 pandemic caused significant disruptions in healthcare services. Foreign body impactions (FBIs), with Eosinophilic Esophagitis (EoE) being one of the leading underlying causes in adults, are some of the most common emergencies and often require endoscopy. The study assesses the impact of COVID-19 on the incidence and outcomes of foreign body impactions (FBIs) requiring endoscopy at Canadian tertiary centres in a single city. Methods: Patients presenting to tertiary care hospital emergency departments in Calgary (March 2019–Feb 2022) for FBI were identified using the AACRS (Alberta Ambulatory Care Reporting System) database using International Classification of Disease (ICD-9 and ICD-10) codes (T178, T181) and provincial diagnostic codes (935.1, 530.4) for a foreign body in the esophagus (530.13 and K20.0). One-way ANOVA (SPSS^® 27.0) analyzed incidence and disease progression across Pre-COVID-19 and COVID-19 years. Results: 759 patients were included in the analysis (274 Pre-COVID-19 (PC: March 2019–Feb 2020), 234 COVID-19 Year 1 (CY1: March 2020–Feb 2021), and 251 COVID-19 Year 2 (CY2: March 2021–Feb 2022)). The mean age remained consistent, with two-thirds being male. Food was the predominant type of FBI (>90%). The incidence of new EoE in EDs declined from PC (60.9%) to CY1 (47.4%) (p < 0.001), while endoscopic resolution remained >96%. Follow-up endoscopies in outpatient settings remained stable (~60%). Non-EoE causes of FBI, including esophagitis and cancer, increased in CY2. The mean ED length of stay rose in CY2, but this was not statistically significant (p = 0.06). Conclusions: This study highlights the resilience of emergent endoscopic care in Calgary during COVID, despite a decline in new EoE diagnoses, which might be due to access barriers. Full article

The increasing deployment of internet of things(IoT) systems across critical domains has broadened the threat landscape, and being the catalyst for a variety of security concerns, including very stealthy slow denial of service (slow DoS) attacks. These exploit the hypertext transfer protocol’s (HTTP) application-layer protocol to either close down service requests or degrade responsiveness while closely mimicking legitimate traffic. Current available datasets fail to capture the more stealthy operational profiles of slow DoS attacks or account for the presence of genuine slow nodes (SN), which are devices experiencing high latency. These can significantly degrade detection accuracy since slow DoS attacks closely emulate SN. This paper addresses these problems by synthesising a realistic HTTP slow DoS dataset derived from a live IoT network, that incorporates both stealth-tuned slow DoS traffic and legitimate SN traffic, with the three main slow DoS variants of slow GET, slow Read, and slow POST being critically evaluated under these network conditions. A limited packet capture (LPC) strategy is adopted which focuses on just two metadata attributes, namely packet length ($lp$) and packet inter-arrival time ($\mathrm{\Delta }t$). Using a resource lightweight decision tree classifier, the proposed model achieves over 96% accuracy while incurring minimal computational overheads. Experimental results in a live IoT network reveal the negative classification impact of including SN traffic, thereby underscoring the importance of modelling stealthy attacks and SN latency in any slow DoS detection framework. Finally, a MPerf (Modelling Performance) is presented which quantifies and balances detection accuracy against processing costs to facilitate scalable deployment of low-cost detection models in resource-constrained IoT networks. This represents a practical solution to improving IoT resilience against stealthy slow DoS attacks whilst pragmatically balancing the resource-constraints of IoT nodes. By analysing the impact of SN on detection performance, a robust reliable model has been developed which can both measure and fine tune the accuracy-efficiency nexus. Full article

Smart Mobile Devices (SMD)—including hardware devices, such as smartphones, tablets, and wearables; the software systems that animate them; and the data-communication infrastructure that connects them—pose increasing sustainability challenges due to their short lifespans, high resource demands, and growing e-waste. While Sustainable Product–Service Systems (S.PSS) have been applied in various sectors to support environmental goals, limited research has addressed their application in the context of SMD. This study aims to explore how S.PSS can be tailored to support sustainability in the SMD sector. For that, it combines a literature review with a multiple-case analysis of seventeen commercial offerings to develop a conceptual framework refined through six expert interviews. Cases were coded using the classical PSS typology and other sector-specific criteria and subsequently clustered in a polarity diagram to identify designable patterns, underpinning the conceptual framework. The study contributes an S.PSS-SMD framework comprising a sector-tailored classification and sixteen archetypal models, operationalized in an archetypal map with potential opportunities. Theoretically, the study offers a sector-grounded operationalization that extends S.PSS design theory to digital product–service ecosystems. It provides a strategic decision aid for designing business models, service bundles, stakeholder roles, and lifecycle responsibilities to pursue win–win environmental and economic sustainability. Full article

Accurate cost estimation during the conceptual and feasibility phase of highway projects is essential for informed decision making by public contracting authorities. Existing approaches often rely on pavement cross-section descriptors, general project classifications, or quantity estimates of major work categories that are not reliably available at the early planning stage, while focusing on one or more key asset categories such as roadworks, bridges or tunnels. This study makes a novel contribution to both scientific literature and practice by proposing the first early-stage highway construction cost estimation model that explicitly incorporates roadworks, interchanges, tunnels and bridges, using only readily available or easily derived geometric characteristics. A comprehensive and practical approach was adopted by developing and comparing models across multiple machine learning (ML) methods, including Multilayer Perceptron-Artificial Neural Network (MLP-ANN), Radial Basis Function-Artificial Neural Network (RBF-ANN), Multiple Linear Regression (MLR), Random Forests (RF), Support Vector Regression (SVR), XGBoost Technique, and K-Nearest Neighbors (KNN). Results demonstrate that the MLR model based on six independent variables—mainline length, service road length, number of interchanges, total area of structures, tunnel length, and number of culverts—consistently outperformed more complex alternatives. The full MLR model, including its coefficients and standardized parameters, is provided, enabling direct replication and immediate use by contracting authorities, hence supporting more informed decisions on project funding and procurement. Full article

This research presents a methodology for the detection of distributed denial-of-service (DDoS) attacks in software-defined networks (SDNs). An SDN was configured using the Mininet simulator, the Open Daylight controller, and a web server, which acted as the target to execute a DDoS attack on the HTTP protocol. The attack tools GoldenEye, Slowloris, HULK, Slowhttptest, and XerXes were used, and two datasets were built using the CICFlowMeter and NTLFlowLyzer flow and feature generation tools, with 424,922 and 731,589 flows, respectively, as well as two independent test datasets. These tools were used to compare their functionalities and efficiency in generating flows and features. Finally, the XGBoost and Random Forest models were evaluated with each dataset, with the objective of identifying the model that provides the best classification result in the detection of malicious traffic. For the XGBoost model, the accuracy results were 99.48% and 97.61%, while for the Random Forest model, better results were obtained with 99.97% and 99.99% using the CIC-Dataset and NTL-Dataset, respectively, in both cases. This allows determining that the Random Forest model outperformed XGBoost in classification, as it achieved the lowest false negative rate of 0.00001 using the NTL-Dataset. Full article

To address the challenge of efficiently identifying and providing early warnings for typical structural damages in small and medium-sized bridges during long-term service, this paper proposes an intelligent monitoring and recognition method based on ultra-weak fiber Bragg grating (UWFBG) array sensing. By deploying UWFBG strain-sensing cables across the bridge, the system enables continuous acquisition and spatial analysis of multi-point strain data. Based on this, a series of experimental scenarios simulating typical structural damages—such as single-slab loading, eccentric loading, and bearing detachment—are designed to systematically analyze strain evolution patterns before and after damage occurrence. While strain distribution maps allow for visual identification of some typical damages, the approach remains limited by reliance on manual interpretation, low recognition efficiency, and weak detection capability for atypical damages. To overcome these limitations, machine learning algorithms are further introduced to extract features from strain data and perform pattern recognition, enabling the construction of an automated damage identification model. This approach enhances both the accuracy and robustness of damage recognition, achieving rapid classification and intelligent diagnosis of structural conditions. The results demonstrate that the integration of the monitoring system with intelligent recognition algorithms effectively distinguishes different types of damage and shows promising potential for engineering applications. Full article

Road traffic accidents remain a leading cause of fatalities worldwide, and the consequences are considerably worsened by delayed detection and emergency response. Although several machine learning-based approaches have been proposed, accident detection systems are not widely deployed, and most existing solutions fail to handle the growing complexity of modern traffic environments. This study introduces Collaborative Intelligence for Road Safety (CIRS), a novel, multi-agent, machine-learning-based framework designed for real-time accident detection, semantic scene understanding, and coordinated emergency response. Each agent in CIRS is designed for a distinct role perception, classification, description, localization, and decision-making, working collaboratively to enhance situational awareness and response efficiency. These agents integrate advanced models: YOLOv11 for high-accuracy accident detection and VideoLLaMA3 for contextual-rich scene description. CIRS bridges the gap between low-level visual analysis and high-level situational awareness. Extensive evaluation on a custom dataset comprising (5200 accident, 4800 nonaccident) frames demonstrates the effectiveness of the proposed approach. YOLOv11 achieves a top-1 accuracy of 86.5% and a perfect top-5 accuracy of 100%, ensuring reliable real-time detection. VideoLLaMA3 outperforms other vision-language models with superior factual accuracy and fewer hallucinations, generating strong results in the metrics of BLEU (0.0755), METEOR (0.2258), and ROUGE-L (0.3625). The decentralized multi-agent architecture of CIRS enables scalability, reduced latency, and the timely dispatch of emergency services while minimizing false positives. Full article

Background: Accurate estimation of mortality is essential for effective public health planning, policymaking, and monitoring of health interventions. In Ghana, multiple data sources are used to estimate mortality, including civil registration systems, household surveys, census data, and health and demographic surveillance systems. This scoping review explores the existing sources of mortality data in Ghana, examining their challenges and opportunities. Methods: Using Arksey and O’Malley’s framework, we identified peer-reviewed and grey literature from Ghana Health Service, Ministry of Health, Ghana Statistical Service, WHO, and the United Nations. We selected studies published between 2000 and 2024 that focused on mortality estimation in Ghana. Data was extracted and synthesized into key themes: data sources, challenges, and opportunities. Results: Six major data sources on mortality were identified: Civil Registration and Vital Statistics (CRVS), census data, Demographic and Health Surveys (DHS), Health and Demographic Surveillance Systems (HDSS), Facility-Based Health Information Systems (HMIS), modeled estimates from the Global Burden of Disease (GBD) and from the United Nations Department of Economic and Social Affairs (UN DESA). Key challenges include under-registration of deaths (CRVS and HMIS), recall bias (DHS, census), limited geographic coverage (HDSS), inconsistencies in cause-of-death classification (HMIS, HDSS), and lack of local geographic coverage (GBD, UN DESA, DHS). Nonetheless, benefits include longitudinal follow-up (HDSS), local coverage and ownership (CRVS, HMIS) and international comparability (GBD, UN DESA, DHS). Conclusions: Mortality estimation in Ghana is supported by diverse but fragmented data sources. Strengthening the CRVS and HMIS systems, integrating multiple data streams, standardizing methodologies, and enhancing institutional partnership are essential steps toward improving data quality and coverage. This review provides recommendations for improvement towards better quality estimations of mortality in Ghana. Full article

Machine learning in natural language processing (NLP) analyzes datasets to make future predictions, but developing accurate models requires large, high-quality, and balanced datasets. However, collecting such datasets, especially for low-resource languages, is time-consuming and costly. As a solution, data augmentation can be used to increase the dataset size by generating synthetic samples from existing data. This study examines the effect of translation-based data augmentation on sentiment analysis using small datasets in three diverse languages: French, German, and Japanese. We use two neural machine translation (NMT) services—Google Translate and DeepL—to generate augmented datasets through intermediate language translation. Sentiment analysis models based on Support Vector Machine (SVM) are trained on both original and augmented datasets and evaluated using accuracy, precision, recall, and F1 score. Our results demonstrate that translation augmentation significantly enhances model performance in both French and Japanese. For example, using Google Translate, model accuracy improved from 62.50% to 83.55% in Japanese (+21.05%) and from 87.66% to 90.26% in French (+2.6%). In contrast, the German dataset showed a minor improvement or decline, depending on the translator used. Google-based augmentation generally outperformed DeepL, which yielded smaller or negative gains. To evaluate cross-lingual generalization, models trained on one language were tested on datasets in the other two. Notably, a model trained on augmented German data improved its accuracy on French test data from 81.17% to 85.71% and on Japanese test data from 71.71% to 79.61%. Similarly, a model trained on augmented Japanese data improved accuracy on German test data by up to 3.4%. These findings highlight that translation-based augmentation can enhance sentiment classification and cross-language adaptability, particularly in low-resource and multilingual NLP settings. Full article

Mapping forest tree species is vital for the habitat assessment, ecosystem services estimation, and implementation of European environmental policies such as the Habitats Directive. This study explores how repeated satellite observations over time, known as multitemporal data, can improve the mapping of tree species in riparian forests. Although many studies have shown that the use of multitemporal data improves tree species classification accuracies, there is a lack of research on how different multitemporal models perform compared to each other. We compared three multitemporal remote sensing approaches using Sentinel-2 imagery to map tree species within the Austrian riparian Natura 2000 site, Salzachauen. Seven tree species (five native and two non-native riparian species) were mapped using random forest models trained on a dataset of 444 validated tree samples. The three multitemporal approaches tested were: (i) multi-date image stacking, (ii) seasonal mean composites, and (iii) spectral–temporal metrics (STMs). The three approaches were compared to twenty single-date image classifications. The multitemporal models achieved 62 to 65% overall accuracy, while the median accuracy of single-date classification was 50% (SD = 6%). The seasonal model obtained the highest overall accuracy (65%), with F1 scores exceeding 73% for four individual species. However, differences among the three multitemporal approaches were not statistically significant. The mapping of native versus non-native riparian species achieved 92% accuracy. We evaluated misclassification patterns of individual species according to the two riparian forest habitats, 91E0* and 91F0, as defined in Annex I of the Habitats Directive. Most omission and commission errors occurred between species within the same habitat type. These findings underline the potential of translating tree species mapping to habitat-type classifications and the need to further explore the capabilities of satellite remote sensing to fill data gaps in Natura 2000 areas. Full article

Background: Chronic mast cell–mediated angioedema (MC-AE) and chronic spontaneous urticaria (CSU) both involve mast cell activation but may differ in long-term systemic outcomes. Limited data exist comparing their comorbidity profiles over extended follow-up. Objective: To compare systemic comorbidities in patients with chronic MC-AE versus CSU using a large, population-based dataset. Methods: We conducted a retrospective matched case–control study using electronic health records from Leumit Health Services, a nationwide Israeli health maintenance organization. Patients diagnosed with chronic MC-AE between 2005 and 2023 (n = 2133) were matched 1:1 by age, sex, and year of diagnosis to patients with CSU (n = 2133). Comorbidities were assessed at diagnosis and after a mean follow-up of 10.2 ± 2.9 years. Odds ratios (ORs) with 95% confidence intervals (CIs) were calculated. Multivariable logistic regression was used to assess the association between medications and MC-AE diagnosis. Results: MC-AE patients exhibited significantly higher baseline rates of hypertension (23.8% vs. 18.5%), ischemic heart disease (5.67% vs. 3.84%), and type 2 diabetes (10.45% vs. 6.42%) compared to CSU. These differences persisted or increased at follow-up, including myocardial infarction (4.13% vs. 2.25%) and chronic kidney disease (4.13% vs. 2.91%). CSU patients had consistently higher rates of atopic dermatitis, viral infections, and herpes zoster. Statin use was inversely associated with MC-AE (adjusted OR = 0.63; 95% CI: 0.44–0.90). Conclusions: Chronic MC-AE is associated with a distinct and sustained cardiometabolic and renal comorbidity burden compared to CSU, supporting its classification as a systemic disease phenotype requiring differentiated long-term care. Full article

Background and Objectives: Professional burnout threatens special education teachers’ well-being and educational service quality through three psychological dimensions: emotional exhaustion, depersonalization, and personal achievement. Limited studies have employed behavioral data analysis to examine burnout patterns in special education and their relationships with demographic factors and contemporary stressors. This study aimed to (1) identify burnout levels among Greek special education teachers, (2) determine demographic risk factors, and (3) examine relationships between burnout dimensions and COVID-19 psychological impact. Materials and Methods: A cross-sectional study surveyed 114 special education teachers from Achaia and Aitoloakarnania prefectures, Greece (response rate: 87.7%), using the Maslach Burnout Inventory–Educators Survey (MBI-ES) and demographic questionnaires. Behavioral data analysis integrates traditional statistics with advanced techniques, including cluster analysis and classification modeling. Results: Four distinct burnout profiles emerged: Low Burnout (36.8%), Moderate Emotional Exhaustion (30.7%), High Risk (21.9%), and Depersonalization-Dominant (10.5%). Overall burnout prevalence was low, with 73.7% showing minimal depersonalization and 67.5% maintaining high personal achievement. Employment status emerged as the strongest predictor of burnout risk. Emotional exhaustion was the primary predictor of COVID-19 psychological impact (r = 0.547, p < 0.001), explaining 29.9% of pandemic-related distress variance. Male substitute teachers demonstrated the highest vulnerability to depersonalization, while experienced female permanent teachers showed resilience patterns. Conclusions: Behavioral data analysis revealed distinct burnout patterns enabling personalized interventions. Emotional exhaustion serves as both a key vulnerability factor and primary intervention target. These findings support targeted approaches to occupational health with implications for educational policy. Limitations include cross-sectional design and regional sampling. Future longitudinal studies should validate these patterns across diverse educational contexts. Full article