Search Results (232)

International organizations have highlighted the importance of consistent and reliable environment, social and governance (ESG) disclosure and metrics to inform business strategy and investment decisions. Greater corporate disclosure is a positive signal to investors who prioritize sustainable investment. In this study, economic and climate sentiment are extracted from the integrated and sustainability reports of the top 40 corporates listed on the Johannesburg Stock Exchange, employing domain-specific natural language processing. The intention is to clarify the complex interactions between climate risk, corporate disclosures, financial performance and investor sentiment. The study provides valuable insights to regulators, accounting professionals and investors on the current state of disclosures and future actions required in South Africa. A time series analysis of the sentiment scores indicates a noticeable change in the corporates’ disclosures from climate-related risks in the earlier years to climate-related opportunities in recent years, specifically in the banking and mining sectors. The trends are less pronounced in sectors with good ESG ratings. An exploratory regression study reveals that climate and economic sentiments contain information that explain stock price movements over the longer term. The results have important implications for asset allocation and offer an interesting direction for future research. Monitoring the sentiment may provide early-warning signals of systemic risk, which is important to regulators given the impact on financial stability. Full article

The influence of seafarers’ emotions on work performance can lead to severe marine accidents. However, research on emotion recognition (ER) of seafarers remains insufficient, and existing studies only deploy single models and disregard the model's uncertainty, which might lead to unreliable recognition. In this paper, a novel fusion framework for seafarer ER is proposed. Firstly, feature-level fusion using Electroencephalogram (EEG) and navigation data collected in a real navigation environment was conducted. Then, calibration is employed to mitigate the uncertainty of the outcomes. Secondly, a weight combination strategy for decision fusion was designed. Finally, we conduct a series of evaluations of the proposed model. The results showed that the average recognition performance across the three emotional dimensions, as measured by accuracy, precision, recall, and F1 score, reaches 85.14%, 84.43%, 86.27%, and 85.33%, respectively. The results demonstrate that the use of physiological and navigation data can effectively identify seafarers' emotional states. Additionally, the fusion model compensates for the uncertainty of single models and enhances the performance of ER for seafarers, which provides a feasible path for the ER of seafarers. The findings of this study can be used to promptly identify the emotional state of seafarers and develop early warnings for bridge systems for shipping companies and help inform policy-making on human factors to enhance maritime safety. Full article

Student dropout remains a critical challenge in higher education, especially within public universities that serve diverse and vulnerable populations. This research presents the design and evaluation of an early warning system based on an XGBoost classifier, trained exclusively on data collected at the time of student enrollment. Using a retrospective dataset of nearly 40,000 first-year students (2014–2024) from a Mexican public university, the model incorporated academic, socioeconomic, demographic, and perceptual variables. The final XGBoost model achieved an AUC-ROC of 0.6902 and an F1-score of 0.6946 for the dropout class, with a sensitivity of 88%. XGBoost was chosen over Random Forest due to its superior ability to detect students at risk, a critical requirement for early intervention. The model flagged 59% of incoming students as high-risk, with considerable variability across academic programs. The most influential predictors included age, high school GPA, conditioned admission, and other family responsibilities and economic constraints. This research demonstrates that early warning systems can transform enrollment data into timely and actionable insights, enabling universities to identify vulnerable students earlier and respond more effectively, allocate support more efficiently, and enhance their efforts to reduce dropout rates and improve student retention. Full article

The COVID-19 pandemic highlighted the need for a scientific framework that enables quantitative assessment and control of airborne infection risks in indoor environments. This study identifies limitations in the traditional Wells–Riley model—specifically its assumptions of perfect mixing and steady-state conditions—and addresses these shortcomings by adopting the REHVA (Federation of European Heating, Ventilation and Air Conditioning Associations) infection risk assessment model. We propose a five-tier risk classification system (Monitor, Caution, Alert, High Risk, Critical) based on two key metrics: the probability of infection (Pₙ) and the event reproduction number (R_event). Unlike the classical model, our approach integrates airborne virus removal mechanisms—such as natural decay, gravitational settling, and filtration—with occupant dynamics to reflect realistic contagion scenarios. Simulations were conducted across 10 representative indoor settings—such as classrooms, hospital waiting rooms, public transit, and restaurants—considering ventilation rates and activity-specific viral emission patterns. The results quantify how environmental variables (ventilation, occupancy, time) impact each setting’s infection risk level. Our findings indicate that static mitigation measures such as mask-wearing or physical distancing are insufficient without dynamic, model-based risk evaluation. We emphasize the importance of incorporating real-time crowd density, occupancy duration, and movement trajectories into risk scoring. To support this, we propose integrating computer vision (CCTV-based crowd detection) and entry/exit counting sensors within a live airborne risk assessment framework. This integrated system would enable proactive, science-driven epidemic control strategies, supporting real-time adaptive interventions in indoor spaces. The proposed platform could serve as a practical tool for early warning and management during future airborne disease outbreaks. Full article

The study is devoted to the construction of an avalanche susceptibility map based on ensemble machine learning algorithms (random forest, XGBoost, LightGBM, gradient boosting machines, AdaBoost, NGBoost) for the conditions of the East Kazakhstan region. To train these models, data were collected on avalanche path profiles, meteorological conditions, and historical avalanche events. The quality of the trained machine learning models was assessed using metrics such as accuracy, precision, true positive rate (recall), and F1-score. The obtained metrics indicated that the trained machine learning models achieved reasonably accurate forecasting performance (forecast accuracy from 67% to 73.8%). ROC curves were also constructed for each obtained model for evaluation. The resulting AUCs for these ROC curves showed acceptable levels (from 0.57 to 0.73), which also indicated that the presented models could be used to predict avalanche danger. In addition, for each machine learning model, we determined the importance of the indicators used to predict avalanche danger. Analysis of the importance of the indicators showed that the most significant indicators were meteorological data, namely temperature and snow cover level in avalanche paths. Among the indicators that characterized the avalanche paths’ profiles, the most important were the minimum and maximum slope elevations. Thus, within the framework of this study, a highly accurate model was built using geospatial and meteorological data that allows identifying potentially dangerous slope areas. These results can support territorial planning, the design of protective infrastructure, and the development of early warning systems to mitigate avalanche risks. Full article

Coseismic landslides are geological events in which slopes, either on the verge of instability or already in a fragile state, experience premature failure due to seismic shaking. On 5 September 2022, an Ms 6.8 earthquake struck Luding County, Sichuan Province, China, triggering numerous landslides that caused severe casualties and property damage. This study systematically interprets 13,717 coseismic landslides in the Luding earthquake’s epicentral area, analyzing their spatial distribution concerning various factors, including elevation, slope gradient, slope aspect, plan curvature, profile curvature, surface cutting degree, topographic relief, elevation coefficient variation, lithology, distance to faults, epicentral distance, peak ground acceleration (PGA), distance to rivers, fractional vegetation cover (FVC), and distance to roads. The analytic hierarchy process (AHP) was improved by incorporating frequency ratio (FR) to address the subjectivity inherent in expert scoring for factor weighting. The improved AHP, combined with the Pearson correlation analysis, was used to identify the dominant controlling factor and assess the landslide susceptibility. The accuracy of the model was verified using the area under the receiver operating characteristic (ROC) curve (AUC). The results reveal that 34% of the study area falls into very-high- and high-susceptibility zones, primarily along the Moxi segment of the Xianshuihe fault and both sides of the Dadu river valley. Tianwan, Caoke, Detuo, and Moxi are at particularly high risk of coseismic landslides. The elevation coefficient variation, slope aspect, and slope gradient are identified as the dominant controlling factors for landslide development. The reliability of the proposed model was evaluated by calculating the AUC, yielding a value of 0.8445, demonstrating high reliability. This study advances coseismic landslide susceptibility assessment and provides scientific support for post-earthquake reconstruction in Luding. Beyond academic insight, the findings offer practical guidance for delineating priority zones for risk mitigation, planning targeted engineering interventions, and establishing early warning and monitoring strategies to reduce the potential impacts of future seismic events. Full article

Grading and providing early warning of traffic congestion density is crucial for the timely coordination and optimization of traffic management. However, current traffic density detection methods primarily rely on historical traffic flow data, resulting in ambiguous thresholds for congestion classification. To overcome these challenges, this paper proposes a traffic density grading algorithm for remote sensing images that integrates adaptive clustering and multi-scale fusion. A dynamic neighborhood radius adjustment mechanism guided by spatial distribution characteristics is introduced to ensure consistency between the density clustering parameter space and the decision domain for image cropping, thereby addressing the issues of large errors and low efficiency in existing cropping techniques. Furthermore, a hierarchical detection framework is developed by incorporating a dynamic background suppression strategy to fuse multi-scale spatiotemporal features, thereby enhancing the detection accuracy of small objects in remote sensing imagery. Additionally, we propose a novel method that combines density analysis with pixel-level gradient quantification to construct a traffic state evaluation model featuring a dual optimization strategy. This enables precise detection and grading of traffic congestion areas while maintaining low computational overhead. Experimental results demonstrate that the proposed approach achieves average precision (AP) scores of 32.6% on the VisDrone dataset and 16.2% on the UAVDT dataset. Full article

(1) Background: Educational systems across the world experienced significant changes during 2020–2022, with potential implications for progress toward Sustainable Development Goal 4 (SDG 4: Quality Education), which aims to ensure inclusive and equitable quality education and promote lifelong learning opportunities for all across European Union member states. Understanding how educational sustainability performance evolved during the pre-pandemic period (2017–2019) and the pandemic-affected period (2020–2022) is essential for developing effective educational policies. (2) Objective: This quantitative comparative study aimed to (i) assess and rank sustainable education developments across EU-27 countries in two periods, Period 1—the pre-pandemic period (2017–2019)—and Period 2—the pandemic-affected period (2020–2022); (ii) identify performance changes between these periods; and (iii) classify countries into performance groups to guide targeted interventions. (3) Methods: Using data from the Eurostat database, we evaluated six key SDG 4 indicators: low-achieving students in reading, mathematics, and science; participation in early childhood education; early school leavers; tertiary educational attainment; adult participation in learning; and adults with basic digital skills. The Multiobjective Optimization based on Ratio Analysis (MOORA) method was used to rank countries and assess sustainable education development. (4) Results: Sweden maintained the highest educational sustainability performance across both periods, while Romania and Bulgaria consistently ranked lowest. Nine countries improved their rankings during the pandemic-affected period, while others maintained stable positions or experienced declines in their rankings. Adult participation in learning showed the greatest variation among the indicators, with top performers, such as Sweden, scoring 0.445 compared to Romania’s 0.051 in Period 2. The proportion of early school leavers decreased from an EU average of 9.0% in Period 1 to 8.3% in Period 2, indicating a positive trend across the study periods. While differences were observed across countries and periods, these should not be interpreted as causally linked to the pandemic alone (5). Conclusions: The performance of educational sustainability varied across EU member states between the two periods, with some countries demonstrating remarkable resilience or improvement, while others declined. These findings underscore the need for targeted educational policies that address specific sustainability weaknesses in individual countries, particularly those in the warning and danger categories. Sweden’s consistent performance offers valuable lessons for educational sustainability, especially during and after major disruptions. Full article

China is one of the countries severely affected by earthquakes, making precise and timely identification of earthquake precursors essential for reducing casualties and property damage. A novel method is proposed that combines a rock acoustic emission (AE) detection technique with deep learning methods to facilitate real-time monitoring and advance earthquake precursor detection. The AE equipment and seismometers were installed in a granite tunnel 150 m deep in the mountains of eastern Guangdong, China, allowing for the collection of experimental data on the correlation between rock AE and seismic activity. The deep learning model uses features from rock AE time series, including AE events, rate, frequency, and amplitude, as inputs, and estimates the likelihood of seismic events as the output. Precursor features are extracted to create the AE and seismic dataset, and three deep learning models are trained using neural networks, with validation and testing. The results show that after 1000 training cycles, the deep learning model achieves an accuracy of 98.7% on the validation set. On the test set, it reaches a recognition accuracy of 97.6%, with a recall rate of 99.6% and an F₁ score of 0.975. Additionally, it successfully identified the two biggest seismic events during the monitoring period, confirming its effectiveness in practical applications. Compared to traditional analysis methods, the deep learning model can automatically process and analyse recorded massive AE data, enabling real-time monitoring of seismic events and timely earthquake warning in the future. This study serves as a valuable reference for earthquake disaster prevention and intelligent early warning. Full article

Early detection of clinical deterioration in emergency department (ED) patients is critical for timely interventions. This study evaluated the predictive performance of the National Early Warning Score (NEWS) parameters using machine learning. We conducted a single-center retrospective observational study including 27,238 adult ED patients admitted to Merano Hospital (Italy) between June 2022 and June 2023. NEWS vital signs were collected at triage, and mortality at 48 h, 7 days, and 30 days was obtained from ED database. Decision tree analysis (CHAID algorithm) was used to identify predictors of mortality; 10-fold cross-validation was applied to avoid overfitting. Mortality was 0.4% at 48 h, 1% at 7 days, and 2.45% at 30 days. For 48-h mortality, oxygen supplementation (FiO₂ >21%) and AVPU = “U” were the strongest predictors, with a maximum risk of 31.6%. For 7-day mortality, SpO₂ was the key predictor, with mortality up to 48.1%. At 30 days, patients with AVPU ≠ A, FiO₂ > 21%, and SpO₂ ≤ 94% had a mortality risk of 66.7%. Decision trees revealed different cut-offs compared to the standard NEWS. This study demonstrated that for ED patients, the NEWS may require some adjustments in both the cut-offs for vital parameters and the methods of collecting these parameters. Full article

In the context of intelligent coal mine safety monitoring, an integrated prediction and early-warning method named MTGNN-Bayesian-IF-DBSCAN (Multi-Task Graph Neural Network–Bayesian Optimization–Isolation Forest–Density-Based Spatial Clustering of Applications with Noise) is proposed to address the challenges of gas concentration prediction and anomaly detection in coal mining faces. The MTGNN (Multi-Task Graph Neural Network) is first employed to model the spatiotemporal coupling characteristics of gas concentration and wind speed data. By constructing a graph structure based on sensor spatial dependencies and utilizing temporal convolutional layers to capture long short-term time-series features, the high-precision dynamic prediction of gas concentrations is achieved via the MTGNN. Experimental results indicate that the MTGNN outperforms comparative algorithms, such as CrossGNN and FourierGNN, in prediction accuracy, with the mean absolute error (MAE) being as low as 0.00237 and the root mean square error (RMSE) maintained below 0.0203 across different sensor locations (T0, T1, T2). For anomaly detection, a Bayesian optimization framework is introduced to adaptively optimize the fusion weights of IF (Isolation Forest) and DBSCAN (Density-Based Spatial Clustering of Applications with Noise). Through defining the objective function as the F1 score and employing Gaussian process surrogate models, the optimal weight combination (w_if = 0.43, w_dbscan = 0.52) is determined, achieving an F1 score of 1.0. By integrating original concentration data and residual features, gas anomalies are effectively identified by the proposed method, with the detection rate reaching a range of 93–96% and the false alarm rate controlled below 5%. Multidimensional analysis diagrams (e.g., residual distribution, 45° diagonal error plot, and boxplots) further validate the model’s robustness in different spatial locations, particularly in capturing abrupt changes and low-concentration anomalies. This study provides a new technical pathway for intelligent gas warning in coal mines, integrating spatiotemporal modeling, multi-algorithm fusion, and statistical optimization. The proposed framework not only enhances the accuracy and reliability of gas prediction and anomaly detection but also demonstrates potential for generalization to other industrial sensor networks. Full article

The Mw 7.8 Gorkha Earthquake of 25 April 2015 triggered over 25,000 landslides across central Nepal, with 4775 events concentrated in Gorkha District alone. Despite substantial advances in landslide susceptibility mapping, existing studies often overlook the compound role of post-seismic rainfall and lack robust spatial validation. To address this gap, we validated an ensemble machine learning framework for co-seismic landslide susceptibility modeling by integrating seismic, geomorphological, hydrological, and anthropogenic variables, including cumulative post-seismic rainfall. Using a balanced dataset of 4775 landslide and non-landslide instances, we evaluated the performance of Logistic Regression (LR), Random Forest (RF), and eXtreme Gradient Boosting (XGBoost) models through spatial cross-validation, SHapley Additive exPlanations (SHAP) explainability, and ablation analysis. The RF model outperformed all others, achieving an accuracy of 87.9% and a Receiver Operating Characteristic (ROC) Area Under the Curve (AUC) value of 0.94, while XGBoost closely followed (AUC = 0.93). Ensemble models collectively classified over 95% of observed landslides into High and Very High susceptibility zones, demonstrating strong spatial reliability. SHAP analysis identified elevation, proximity to fault, peak ground acceleration (PGA), slope, and rainfall as dominant predictors. Notably, the inclusion of post-seismic rainfall substantially improved recall and F1 scores in ablation experiments. Spatial cross-validation revealed the superior generalizability of ensemble models under heterogeneous terrain conditions. The findings underscore the value of integrating post-seismic hydrometeorological factors and spatial validation into susceptibility assessments. We recommend adopting ensemble models, particularly RF, for operational hazard mapping in earthquake-prone mountainous regions. Future research should explore the integration of dynamic rainfall thresholds and physics-informed frameworks to enhance early warning systems and climate resilience. Full article

Background/Objective: In-hospital cardiac arrest has high incidence and poor survival rates, posing a significant healthcare challenge. It is important to intervene in the hours before the cardiac arrest to prevent poor outcomes. The modified early warning score (MEWS) is a validated tool for identifying a deteriorating patient. It is an aggregate of vital signs and level of consciousness. We retrospectively evaluated MEWS for trends that might predict patient outcomes. Methods: We performed a single-center, one-year, retrospective study. A comprehensive review was conducted for patients aged 18 years and above who experienced a cardiac arrest. Cases that occurred within an intensive care unit, emergency department, during a procedure, or outside the hospital were excluded. A total of 87 cases met our predefined inclusion criteria. We collected data at 12 h, 6 h and 1 h time periods prior to the cardiac arrest. A trend analysis using a linear model with analysis of variance with Bonferroni correction was performed. Results: Out of 87 patients included in the study, 59 (67.8%) had an immediate return of spontaneous circulation (ROSC). Among those who achieved ROSC, 41 (69.5%) died during the admission. Only 20.7% of the patients that sustained a cardiac arrest survived to discharge. A significant increase in the average MEWS was noted from the 12 h period (MEWS = 3.95 ± 2.4) to the 1 h period (MEWS = 5.98 ± 3.5) (p ≤ 0.001) and the 6 h period (4.65 ± 2.6) to the 1 h period (5.98 ± 3.5) (p = 0.023) prior to cardiac arrest. Conclusions: An increase in the MEWS may be a valuable tool in identifying at-risk patients and provides an opportunity to intervene at least 6 h before a cardiac arrest event. Further research is needed to validate the results of our study. Full article

While prior work has explored learner behavior using learning management systems (LMS) data, few studies provide week-level clustering validated against external engagement labels. To understand and assist students in online learning platforms and environments, this study presents a week-level engagement profiling framework for e-learning environments, utilizing K-means clustering and label-based validation. Leveraging log data from 127 students over a 13-week course, 44 activity-based features were engineered to classify student engagement into high, moderate, and low levels. The optimal number of clusters (k = 3) was identified using the elbow method and assessed through internal metrics, including a silhouette score of 0.493 and R² of 0.80. External validation confirmed strong alignment with pre-labeled engagement levels based on activity frequency and weighting. The clustering approach successfully revealed distinct behavioral patterns across engagement tiers, enabling a nuanced understanding of student interaction dynamics over time. Regression analysis further demonstrated a significant association between engagement levels and academic performance, underscoring the model’s potential as an early warning system for identifying at-risk learners. These findings suggest that clustering based on LMS behavior offers a scalable, data-driven strategy for improving learner support, personalizing instruction, and enhancing retention and academic outcomes in digital education settings such as MOOCs. Full article

Background/Objectives: Hospitalized peripartum patients who later decompensate and require an upgrade to the intensive care unit (ICU) may have an increased risk for poor outcomes. Most of the literature regarding the need for ICU involves Modified Early Warning Scores in already hospitalized patients or the evaluation of specific comorbid conditions or diagnoses. This systematic review and meta-analysis aimed to assess the differences in clinical scores at admission among adult peripartum patients to identify the later need for ICU. Methods: We systematically searched Ovid-Medline, PubMed, EMBASE, Web of Science and Google Scholar for randomized and observational studies of adult patients ≥18 years of age who were ≥20 weeks pregnant or up to 40 days post-partum, were admitted to the wards from the emergency department and later required critical care services. The primary outcome was the Sequential Organ Failure Assessment (SOFA) score. Secondary outcomes included other clinical scores, the hospital length of stay (HLOS) and mortality. The Newcastle–Ottawa Scale was utilized to grade quality. Descriptive analyses were performed to report demographic data, with means (±standard deviation [SD]) for continuous data and percentages for categorical data. Random-effects meta-analyses were performed for all outcomes when at least two studies reported a common outcome. Results: Seven studies met the criteria, with a total of 1813 peripartum patients. The mean age was 27.2 (±2.36). Patients with ICU upgrades were associated with larger differences in mean SOFA scores. The pooled difference in means was 2.76 (95% CI 1.07–4.46, p < 0.001). There were statistically significant increases in Sepsis in Obstetrics Scores, APACHE II scores, and HLOS in ICU upgrade patients. There was a non-significantly increased risk of mortality in ICU upgrade patients. There was high overall heterogeneity between patient characteristics and management in our included studies. Conclusions: This systematic review and meta-analysis demonstrated higher SOFA or other physiologic scores in ICU upgrade patients compared to those who remained on the wards. ICU upgrade patients were also associated with a longer HLOS and higher mortality compared with control patients. Full article