Search Results (5,135)

This paper presents a comprehensive analysis of artificial intelligence-powered automated grading systems (AI AGSs) in STEM education, systematically examining their algorithmic foundations, mathematical modeling approaches, and quantitative evaluation methodologies. AI AGSs enhance grading efficiency by providing large-scale, instant feedback and reducing educators’ workloads. Compared to traditional manual grading, these systems improve consistency and scalability, supporting a wide range of assessment types, from programming assignments to open-ended responses. This paper provides a structured taxonomy of AI techniques including logistic regression, decision trees, support vector machines, convolutional neural networks, transformers, and generative models, analyzing their mathematical formulations and performance characteristics. It further examines critical challenges, such as user trust issues, potential biases, and students’ over-reliance on automated feedback, alongside quantitative evaluation using precision, recall, F1-score, and Cohen’s Kappa metrics. The analysis includes feature engineering strategies for diverse educational data types and prompt engineering methodologies for large language models. Lastly, we highlight emerging trends, including explainable AI and multimodal assessment systems, offering educators and researchers a mathematical foundation for understanding and implementing AI AGSs into educational practices. Full article

Traditional assessment methods in physical education often emphasize final grades, lacking real-time monitoring and feedback during the learning process. To address this limitation and enhance the formative evaluation of student performance, this study proposes a real-time assessment system for Baduanjin instruction in physical education, utilizing a commercially available inertial measurement unit-based motion capture device. The system was developed in four stages. First, a dataset was created by recruiting 20 university students and one expert physical education instructor. Participants were asked to perform standardized Baduanjin routines while wearing wireless inertial measurement unit sensors on key body joints. The collected kinematic data, sampled at 100 Hz, included joint angles and movement trajectories. Second, preprocessing and feature extraction techniques were applied to the raw data to construct a labeled dataset for training. Third, supervised machine learning algorithms were used to build models for motion type recognition and motion accuracy evaluation. Model performance was assessed using cross-validation and compared with expert evaluations. Finally, a user-facing formative assessment system was developed and tested in a controlled classroom environment. The system demonstrated a high motion recognition accuracy of 99.77%, and the correlation coefficient between system-assessed motion accuracy and expert ratings exceeded 0.80, indicating strong validity. The results demonstrate that the formative assessment system built on inertial measurement unit is appropriate for the Baduanjin physical education. Full article

In online communities, polarization refers to the phenomenon in which individuals become more divided and extreme in their opinions due to their exposure to specific content. In this paper, we present a network-based framework for evaluating polarization levels in Online Social Networks (OSNs). Starting from a dataset of comments, our framework creates a network of user interactions and leverages the Louvain algorithm, the Rao’s Quadratic Entropy, and ego networks to assess the polarization level of communities and the most influential users. To test our framework, we leveraged a dataset of tweets about climate change. After performing Extraction, Transformation and Loading activities on the dataset, we evaluated its labels, identified communities, and analyzed their polarization level and that of the most influential users. We also analyzed the ego networks of believers and deniers and the aggressiveness of the corresponding tweets. Our analysis revealed the existence of polarized communities and homophily among the most influential users. It also showed that the type of communication used to disseminate information influences the polarization level of both communities and individual users. These results demonstrate our framework’s ability to support the polarization analysis in OSNs. Full article

Proportional symbol maps are a frequently used method of thematic cartography. Using an intuitive principle—the larger, the more—provides a simple and precise way of visualizing quantity in maps using geographic information systems (GIS). However, none of the current GIS software provides a proper map legend that could be used to interpret exact phenomenon quantity values from the map in reverse. Cartographers have been designing value scales manually for such a possibility of interpretation. Eventually, they preferred to resign to the accuracy of the interpretation and use the legend offered by the software. The paper describes the development of an easy-to-use online value scale generator for static maps, aiming to eliminate the time-consuming process to make map design more efficient while preserving the precision of cartographic visualization and its subsequent interpretation. The tool consists of a free web platform performing all necessary calculations and rendering an appropriate value scale based on user-defined input parameters. This functionality is performed for most typically used symbol shapes as well as for custom-design shapes provided by the user in SVG vector graphics. The output is then returned in a vector SVG and PDF file format to be used directly in a map legend or possibly edited in graphic software before such a step. The presented tool is therefore independent of which software was used for map design. Within the research, two user experiments were performed to compare generated value scales with simple legends generated in GIS and to gather insights from cartography experts. Full article

Lubricating oils are utilised in equipment and machinery to reduce friction and enhance material utilisation. The utilisation of oil leads to an increase in its thickness and density over time. Current methods for assessing oil life are slow, expensive, and complex, and often only applicable in laboratory settings and unsuitable for real-time or field use. This leads to unexpected equipment failures, unnecessary oil changes, and economic and environmental losses. A comprehensive review of the extant literature revealed no studies and no national or international patents on neural network algorithm-based oil life modelling and classification using green sensors. In order to address this research gap, this study, for the first time in the literature, provides a green conductivity sensor with high-accuracy prediction of oil life by integrating real-time field measurements and artificial neural networks. This design is based on analysing resistance change using a relatively low-cost, three-dimensional, eco-friendly sensor. The sensor is characterised by its simplicity, speed, precision, instantaneous measurement capability, and user-friendliness. The MLP and LVQ algorithms took as input the resistance values measured in two different oil types (diesel, bench oil) after 5–30 h of use. Depending on their degradation levels, they classified the oils as ‘diesel’ or ‘bench oil’ with 99.77% and 100% accuracy. This study encompasses a sensing system with a sensitivity of 50 µS/cm, demonstrating the proposed methodologies’ efficacy. A next-generation decision support system that will perform oil life determination in real time and with excellent efficiency has been introduced into the literature. The components of the sensor structure under scrutiny in this study are conducive to the creation of zero waste, in addition to being environmentally friendly and biocompatible. The developed three-dimensional green sensor simultaneously detects physical (resistance change) and chemical (oxidation-induced polar group formation) degradation by measuring oil conductivity and resistance changes. Measurements were conducted on simulated contaminated samples in a laboratory environment and on real diesel, gasoline, and industrial oil samples. Thanks to its simplicity, rapid applicability, and low cost, the proposed method enables real-time data collection and decision-making in industrial maintenance processes, contributing to the development of predictive maintenance strategies. It also supports environmental sustainability by preventing unnecessary oil changes and reducing waste. Full article

The clinical case presented demonstrates how a person living with type 2 diabetes and treated with insulin reuses the same pen needle several times to save money and performs an incorrect maneuver while screwing the needle, which breaks, remains stuck at the end of the pen, and causes loss of insulin during subsequent use. The findings in this case study are observed in many others in clinical practice but have only been sporadically published. Who is responsible for incorrect injections? Indeed, health workers, diabetic patients, and all the other actors involved in diabetes care and insulin utilization share responsibility. Recommendations and guidelines are not enough to fill this gap. Moreover, not all healthcare providers (HCPs) know or adhere to them. It is observed daily that more than half of insulin users make mistakes that affect glycemic control, increase the risk of complications, and reduce the quality of life of people living with diabetes, who, by a rough estimate, make up a population of over 100 million in the world. This case study offers us the opportunity to briefly review the literature on the most common errors made during insulin injection technique and, therefore, consider how necessary it is to promote structured and coordinated actions among various actors to promote the culture of therapeutic education. Full article

This study presents and evaluates three sampling strategies to characterize electricity demand in regions of Colombia with limited metering infrastructure. These areas lack Advanced Metering Infrastructure (AMI), relying instead on traditional monthly consumption records. The objective of the research is to obtain user samples that are representative of the original population and logistically efficient, in order to support energy planning and decision-making. The analysis draws on five years of historical data from 2020 to 2024. It includes monthly energy consumption, geographic coordinates, customer classification, and population type, covering over 500,000 users across four subregions of operation determined by the region grid operator: North, South, Center, and East. The proposed methodologies are based on Shannon entropy, consumption-based probabilistic sampling, and Kullback–Leibler divergence minimization. Each method is assessed for its ability to capture demand variability, ensure representativeness, and optimize field deployment. Representativeness is evaluated by comparing the differences in class proportions between the sample and the original population, complemented by the Pearson correlation coefficient between their distributions. Results indicate that entropy-based sampling excels in logistical simplicity and preserves categorical diversity, while KL divergence offers the best statistical fit to population characteristics. The findings demonstrate how combining information theory and statistical optimization enables flexible, scalable sampling solutions for demand characterization in under-instrumented electricity grids. Full article

Background and Aims: Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) delay gastric emptying, raising concerns about periprocedural safety in elective endoscopy. We aimed to evaluate the association between pre-procedural GLP-1 RA use and post-procedural complications such as aspiration pneumonia. Methods: In this retrospective cohort study, adults (18–89 years) undergoing outpatient esophagogastroduodenoscopy or colonoscopy within the HCA Healthcare network from 1 July 2021 to 31 March 2024 were identified. Patients were classified as GLP-1 RA users (n = 953) or non-users (n = 3289) based on home medication records. Primary outcomes included aspiration, post-procedural oxygen requirement, hypotension, hospitalization, ICU admission, length of stay, and all-cause inpatient mortality. Multivariable logistic and negative-binomial regression models, incorporating an interaction term for anesthesia type, were adjusted for age, sex, body mass index, ASA class, and key comorbidities. Results: No aspiration events were reported in either group. GLP-1 RA use was associated with lower odds of post-procedural oxygen requirement (OR 0.43, 95% CI 0.25–0.76), hospitalization (OR 0.73, 95% CI 0.39–1.36), and mortality (0.1 vs. 0.9%, p = 0.014), and a shorter hospital stay (IRR 0.54, 95% CI 0.40–0.71). Rates of hypotension and ICU admission were similar between both groups. In anesthesia-stratified analysis among GLP-1 RA users, those receiving MAC/MS had higher odds of hospitalization compared with GA (OR 1.87, 95% CI 1.23–2.85, p = 0.003), whereas other outcomes were not significant. Conclusions: Pre-procedural GLP-1 RA therapy was not associated with increased peri-procedural complications. Although hospitalization was more frequent with MAC/MS, this difference did not extend to other clinically significant outcomes. Further prospective studies are needed to clarify the clinical implications of anesthesia choice. Full article

Nowadays, various technologies exist with differing rendering performance for interactive web maps. These maps are consumed on devices with varying capabilities; therefore, choosing the best-performing library for a dataset is emphasized. Unlike existing research, this study presents a comparative analysis on libraries’ native performance for rendering large amounts of GeoJSON vector data, partially extracted from OpenStreetMap (OSM). Four libraries were analyzed. Results showed that regardless of feature types, Leaflet and OpenLayers excelled for features up to 10,000. Up to 5000 points, these two were the fastest, above which the libraries’ performance converged. For 50,000 or more, Mapbox GL JS rendered them the quickest, followed by OpenLayers, MapLibre GL JS and Leaflet. For up to 50,000 lines and 10,000 polygons, Leaflet and OpenLayers were the fastest in all scenarios. For 100,000 lines, OpenLayers was almost twice as fast as the others, while Mapbox rendered 50,000 polygons the quickest. The performance of Leaflet and OpenLayers scales with the increasing feature quantities, yet for Mapbox and MapLibre, any performance impact is offset to 1000 features and beyond. Slow initalization of map elements makes Mapbox and MapLibre less suitable for rapid rendering of small feature quantities. Other behavioural differences affecting user experience are also explored. Full article

Spherical Adsorptive Carbon (SAC), a type of oral sorbent, is prescribed to chronic kidney disease (CKD) patients to remove uremic toxins. However, evidence regarding its effectiveness in delaying chronic kidney disease (CKD) progression remains insufficient. We aimed to evaluate the impact of SAC on CKD progression in patients with CKD stage 3 or higher using nationwide data. In this retrospective cohort study, we included patients diagnosed with CKD stage ≥3 from the Korea National Health Insurance System database between January 2020 and December 2022. Outcomes were compared between SAC users (N = 1289) and non-users (N = 1289) after 1:1 propensity score matching (PSM). After PSM, the time from index date to end-stage kidney disease (ESKD) was significantly longer in the SAC user group compared to the non-user group (246.8 days vs. 118.6 days, p < 0.001). In Cox regression analysis, the risk of ESKD was significantly lower in the SAC group (HR = 0.37, 95% CI: 0.29–0.48). However, the risk of dialysis initiation did not show a significant difference between the two groups (HR = 0.83, 95% CI: 0.27–2.59). This nationwide cohort study suggests that SAC treatment may delay progression from CKD stage 3 to ESKD, although it did not significantly reduce the risk of dialysis initiation. Full article

Data is needed to understand local air quality, reduce exposure, and mitigate the negative impacts on human health. Measuring local air quality often requires a hybrid monitoring approach consisting of the national air monitoring network and one or more networks of air sensors. However, it can be challenging to combine this data to produce a consistent picture of air quality, largely because sensor data is produced in a variety of formats. Users may have difficulty reformatting, performing basic quality control steps, and using the data for their intended purpose. We developed an R-Shiny application that allows users to import text-based air sensor data, describe the format, perform basic quality control, and export the data to standard formats through a user-friendly interface. Format information can be saved to speed up the processing of additional sensors of the same type. This tool can be used by air quality professionals (e.g., state, local, Tribal air agency staff, consultants, researchers) to more efficiently work with data and perform further analysis in the Air Sensor Network Analysis Tool (ASNAT), Google Earth or Geographic Information System (GIS) programs, the Real Time Geospatial Data Viewer (RETIGO), or other applications they already use for air quality analysis and management. Full article

As network environments become increasingly dynamic and users’ Quality of Service (QoS) demands grow more diverse, efficient and adaptive routing strategies are urgently needed. However, traditional routing strategies suffer from limitations such as poor adaptability to fluctuating traffic, lack of differentiated service handling, and slow convergence in complex network scenarios. To this end, we propose a routing strategy based on multi-agent deep deterministic policy gradient for differentiated QoS services (RS-MADDPG) in a software-defined networking (SDN) environment. First, network state information is collected in real time and transmitted to the control layer for processing. Then, the processed information is forwarded to the intelligent layer. In this layer, multiple agents cooperate during training to learn routing policies that adapt to dynamic network conditions. Finally, the learned policies enable agents to perform adaptive routing decisions that explicitly address differentiated QoS requirements by incorporating a custom reward structure that dynamically balances throughput, delay, and packet loss according to traffic type. Simulation results demonstrate that RS-MADDPG achieves convergence approximately 30 training cycles earlier than baseline methods, while improving average throughput by 3%, reducing latency by 7%, and lowering packet loss rate by 2%. Full article

Polymer recycling is challenging due to practical classification difficulties. Even when the polymer matrix is identified, the presence of various polymeric composites complicates their accurate classification. In this study, Fourier-transform infrared spectroscopy (ATR-FTIR) was used in combination with artificial neural networks (ANNs) to quantitatively predict the mineral filler content in polypropylene (PP) composites. Calibration curves were developed to correlate ATR-FTIR spectral features (600–1700 cm⁻¹) with the concentration (wt.%) of three mineral fillers: talc (PP-Talc), calcium carbonate (PP-CaCO₃), and glass fiber (PP-GF). ANN models developed in MATLAB 2024a achieved prediction errors below 7.5% and regression coefficients (R²) above 0.98 for all filler types. The method was successfully applied to analyze a commercial recycled pellet, and its predictions were validated by X-ray fluorescence (XRF) and energy-dispersive X-ray spectroscopy (EDX). This approach provides a simple, rapid, and non-destructive tool for non-expert users to identify both the type and amount of mineral filler in recycled polymer materials, thereby reducing misclassification in their commercialization or quality control in industrial formulations. Full article

The proliferation of fake profiles on social media presents critical cybersecurity and misinformation challenges, necessitating robust and scalable detection mechanisms. Such profiles weaken consumer trust, reduce user engagement, and ultimately harm brand reputation and platform credibility. As adversarial tactics and synthetic identity generation evolve, traditional rule-based and machine learning approaches struggle to detect evolving and deceptive behavioral patterns embedded in dynamic user-generated content. This study aims to develop an AI-driven, multi-modal deep learning-based detection system for identifying fake profiles that fuses textual, visual, and social network features to enhance detection accuracy. It also seeks to ensure scalability, adversarial robustness, and real-time threat detection capabilities suitable for practical deployment in industrial cybersecurity environments. To achieve these objectives, the current study proposes an integrated AI system that combines the Robustly Optimized BERT Pretraining Approach (RoBERTa) for deep semantic textual analysis, ConvNeXt for high-resolution profile image verification, and Heterogeneous Graph Attention Networks (Hetero-GAT) for modeling complex social interactions. The extracted features from all three modalities are fused through an attention-based late fusion strategy, enhancing interpretability, robustness, and cross-modal learning. Experimental evaluations on large-scale social media datasets demonstrate that the proposed RoBERTa-ConvNeXt-HeteroGAT model significantly outperforms baseline models, including Support Vector Machine (SVM), Random Forest, and Long Short-Term Memory (LSTM). Performance achieves 98.9% accuracy, 98.4% precision, and a 98.6% F1-score, with a per-profile speed of 15.7 milliseconds, enabling real-time applicability. Moreover, the model proves to be resilient against various types of attacks on text, images, and network activity. This study advances the application of AI in cybersecurity by introducing a highly interpretable, multi-modal detection system that strengthens digital trust, supports identity verification, and enhances the security of social media platforms. This alignment of technical robustness with brand trust highlights the system’s value not only in cybersecurity but also in sustaining platform credibility and consumer confidence. This system provides practical value to a wide range of stakeholders, including platform providers, AI researchers, cybersecurity professionals, and public sector regulators, by enabling real-time detection, improving operational efficiency, and safeguarding online ecosystems. Full article

This research analyzes the impact of crime news on users’ digital engagement on social networks. Specifically, this study reviews the influence of presentation format, crime details, and news or discursive values on people’s interaction with media content. To achieve the study’s objective, 1000 posts from the social network Instagram about crimes in the main media outlets of Colombia, Mexico, Paraguay, and Ecuador were reviewed. Content analysis was employed to code the variables, while negative binomial regression models were used to assess their impact on engagement, measured through likes and comments received on the posts. The findings show that shorter videos and image collections generate more engagement than other formats, while the type of crime did not show significant differences in interaction, suggesting a possible normalization of violence among the analyzed viewers. Among the news values, inseparability had a positive effect on engagement, while consonance and references to elite figures demonstrated negative effects. The original value of this research lies in empirically verifying how the characteristics of crime news influence engagement, providing relevant information for understanding the interaction between electronic media and the perception of criminality among Latin American viewers. Full article