Search Results (154)

Predicting student performance has emerged as an essential element of contemporary learning assessment, allowing educational organizations to determine problematic students and offer early intellectual assistance. Many machine learning (ML) methodologies prioritize predicted accuracy at the expense of interpretability and practical insights. This paper provides a framework for predicting student performance that is both risk aware and explainable utilizing a chaotic educational competition optimizer (ECO) in conjunction with a support vector classifier (SVC) to overcome existing challenges. The ECO serves as a metaheuristic feature selection technique for selecting the most significant features from a multivariate educational dataset consisting of 1195 students and 29 behavioral, demographic, and academic characteristics. Experimental findings demonstrate that ECO effectively condenses the feature space to 11 essential indications and improves generalization of model while maintaining classification robustness. Utilizing the chosen features, the ECO–SVC model attains a complete classification accuracy of 87.03%, with F1-scores of 0.92, 0.69, and 0.82 for high-, medium-, and low-performance student categories, respectively, surpassing other benchmark ML methods. The proposed framework incorporates explainable artificial intelligence (XAI) to improve transparency by utilizing local explanations and permutation-driven feature significance. The XAI research verifies that institutional support, learner engagement, and previous academic success are the most important contributing factors to predictive results. Notably the ECO functions as a classifier-independent feature selection mechanism; however, the support vector classifier (SVC) is adopted in this study due to its strong generalization capability and effectiveness in exploiting the optimized feature space. The findings are analyzed using a semiotic-linguistic framework, wherein certain qualities are correlated with symbolic, indexical, and temporal educational signs, converting numerical significance into substantive pedagogical insights. Furthermore, an initial academic risk profile strategy is established by utilizing SVC decision confidence and elucidating feature contributors. The consequent risk ratings accurately categorize students into low-, medium-, and high-risk categories, facilitating the detection of at-risk learners beyond mere final score assessment. The proposed risk-aware and explainable ECO–SVC framework enhances learning outcomes assessment by integrating interpretability, high accuracy, and proactive academic reasoning, rendering it suitable for real-life educational decision-support systems. Full article

The performance of single-photon avalanche diodes (SPADs) is highly dependent on the operating temperature, while traditional SPAD models neglect the self-heating effect induced by avalanche current during long-term device operation, leading to insufficient prediction accuracy. This paper proposes an electro-thermal coupled SPAD simulation model that self-consistently integrates the transient thermal effects of the avalanche process with temperature-dependent electrical parameters, including junction capacitance, breakdown voltage, impact ionization coefficients, and Shockley–Read–Hall (SRH) recombination rates. The complete electro-thermal coupled model is constructed based on Sentaurus-TCAD thermal simulation and Virtuoso circuit simulation and implemented via the Verilog-A language. Simulation results demonstrate that after the device operates for 100 μs under repeated avalanche-quenching processes, the self-heating effect causes a 0.34 V shift in breakdown voltage, increases the device dead time by 3.34 ps, and simultaneously reduces the photon detection probability and elevates the dark count rate. This study conducts a systematic investigation into the performance degradation mechanism of SPAD devices induced by the self-heating effect, laying a theoretical foundation at the device self-heating level for subsequent research on the electrothermal interaction between quenching circuits and device bodies. Full article

The rapid expansion of market-driven software product development has led to the increasing use of User-Generated Content (UGC), such as mobile application user reviews, as a valuable source of requirements. However, unlike the traditional requirements engineering (RE) process, data-driven RE introduces several challenges, particularly in requirements elicitation and prioritization. Traditional requirements prioritization techniques typically rely on stakeholders’ involvement; however, in data-driven and market-driven development contexts, explicit stakeholders are often absent. Thus, we propose a DAta-driven Requirements Prioritization (DARP) framework that integrates Natural Language Processing (NLP), topic modeling, and Large Language Models (LLMs) to automate requirements prioritization in a data-driven development context. The proposed framework utilizes BERTopic to identify latent topics in user reviews and incorporates Hierarchical Density-Based Spatial Clustering of Applications with Noise (HDBSCAN) to group semantically related requirements. The proposed framework introduces a robust and automated prioritization applied to mobile app reviews. The scope of the proposed framework is user-perspective prioritization. Our objective is to detect insights from app reviews to reflect the voice of the customer. The results indicate that leveraging NLP and topic modeling techniques provides an effective data-driven approach to requirements prioritization. Full article

Lightning-induced breaking accidents of medium-voltage insulated conductors pose a serious threat to the safety of distribution networks, and the key cause lies in the establishment and sustained combustion of the power-frequency follow-current arc after lightning overvoltage breakdown. This paper systematically investigates the formation mechanism and critical conditions of power-frequency follow-current arcs using combined simulation and experimental approaches. Based on the streamer discharge theory, a lightning breakdown model was established and combined with the arc energy balance equation, revealing that the establishment of power-frequency follow-current arcs is essentially determined by the post-breakdown energy competition process. The simulation results show that the required anode electric field strength for lightning breakdown is not less than 3 kV/mm. When the power-frequency voltage reaches 10 kV, Joule heating of the arc continuously exceeds heat dissipation loss, enabling restrike after zero-crossing and sustaining stable burning. Experiments verified this voltage threshold and further revealed that the arc establishment rate exhibits nonlinear growth with increasing power-frequency voltage, exceeding 90% at power-frequency voltages ≥ 10 kV. The study also reveals that increased gap distance reduces the arc establishment rate, while the introduction of insulators can enhance it by approximately 20%. This study clarifies the energy criterion for power-frequency follow-current arc establishment and the influence patterns of key parameters, providing theoretical basis and engineering reference for lightning protection design and arc suppression in medium-voltage insulated lines. Full article

This paper suggests a novel approach based on machine learning (ML) and deep learning (DL) for medical image classification in a fast and accurate manner. The proposed method merges the strengths of the convolutional neural network (CNN) using the VGG19 model for feature extraction with an artificial neural network (ANN) classifier for medical dataset classification. The suggested model is improved by applying the slime mold algorithm (SMA) to the task of feature selection and the particle swarm optimization (PSO) approach to optimize the ANN classifier. PSO is a crucial component in neural network design to optimize the ANN setup and hyperparameters. Through adjustments to the bias and weight parameters, the PSO approach enhances the ANN method’s ability to classify medical images. The experiments were conducted on the LC25000 histopathological dataset, which comprises 25,000 histopathological images of lung and colon cancer tissue, partitioned into five classes, each with 5000 images: lung benign tissue, lung adenocarcinoma, lung squamous cell carcinoma, colon adenocarcinoma, and colon benign tissue. The results demonstrated that the suggested model (CNN-PSO-ANN) does better at illness detection than ANN alone. The proposed model is evaluated utilizing several metrics, like accuracy, RMSE, and MAE. The accuracy rate is 94.1% when ANN is utilized independently, while the percentage increases to 98.8% when PSO is employed with the ANN. Additionally, the proposed model is compared with other medical data classification systems that utilize PSO and neural networks. The proposed model (CNN-PSO-ANN) performed better than the other models. With the suggested CNN-PSO-ANN model, diseases, especially cancer, can be found and treated earlier and better. Full article

Because of the substantial class disparity and the intricate interactions between academic, behavioral, and socioeconomic characteristics, anticipating student academic performance and dropout rates continues to be a major issue for institutions of higher learning. To improve the dependability and credibility of multiclass student outcome prediction, this study suggests a strong, multi-objective, and uncertainty-aware predictive framework that combines the Random Forest (RF) classifier with Holistic Swarm Optimization (HSO). The suggested method creates a multi-objective optimization problem that simultaneously maximizes macro F1-score, controls model complexity, and lessens inter-class performance disparity. Thereby, the model promotes fairness across student outcome categories, in contrast to traditional optimization strategies that only concentrate on predictive accuracy. Furthermore, by utilizing ensemble-based probability dispersion, the framework integrates uncertainty-aware prediction, making it possible to identify high-risk students with different degrees of confidence to assist practical academic interventions. According to the results of experiments, the suggested HSO-RF framework greatly reduces the performance gap between outcome classes while achieving the best overall predictive performance, reaching an accuracy of 77.74%, a macro F1-score of 0.69, and a weighted F1-score of 0.76. The analysis shows that academic, socioeconomic, and administrative characteristics serve as significant markers of student motivation, stability, and vulnerability in addition to computational benefits. The suggested architecture advances appropriate and trustworthy educational data mining and offers a dependable decision-support tool for early warning systems. Full article

Evaluating innovation and optimising its role in the inventions is fundamental for applied research, that requires planning the use of available resources. Traditional assessment approaches often miss to capture how innovation stagnates between the ideation and prototyping phases (the Valley of Death), and to learn how innovation emerges from intermediate-steps contributed by individuals. This paper focuses on tracing innovation as an approach enabling mapping of pathways of intermediate-steps and opportunities for valorising unplanned outcomes. We adopt a qualitative case study to explore how innovation pathways can be conceptualised through technological readiness levels. The operational settings of an EU-funded project defined the boundaries of the study. A network analysis explored relationships among themes that emerged from respondents involved in the activities, following an inductive approach to derive themes from data. Findings indicate that intermediate innovation steps, including failures, are viewed as cumulative contributions to novelty. Their documentation is seen as an investment for unlocking latent value embedded in distributed knowledge. Within this scope, we outline a blockchain-based knowledge graph as a proof-of-concept for tracing cumulative contributions, identifying breakthroughs leading to technological maturity and supporting generation of hypothesis grounded on experimental trials. As a result, we suggest that paths recombining prior knowledge into novelty encode latent value that can be interpreted as a function of the network topology, and propose a conceptual framework for analysing value by means of information theory metrics applicable to innovation graphs. Full article

Large volumes of images shared on social media have made image captioning an important tool for social hotspot identification and public opinion monitoring, where accurate visual–language alignment is essential for reliable analysis. However, existing image captioning models based on BLIP-2 (Bootstrapped Language–Image Pre-training) often struggle with complex, context-rich, and socially meaningful images in real-world social media scenarios, mainly due to insufficient cross-modal interaction, redundant visual token representations, and an inadequate ability to capture multi-scale semantic cues. As a result, the generated captions tend to be incomplete or less informative. To address these limitations, this paper proposes ECMA (Enhanced Cross-Modal Attention), a lightweight module integrated into the Querying Transformer (Q-Former) of BLIP-2. ECMA enhances cross-modal interaction through bidirectional attention between visual features and query tokens, enabling more effective information exchange, while a multi-scale visual aggregation strategy is introduced to model semantic representations at different levels of abstraction. In addition, a semantic residual gating mechanism is designed to suppress redundant information while preserving task-relevant features. ECMA can be seamlessly incorporated into BLIP-2 without modifying the original architecture or fine-tuning the vision encoder or the large language model, and is fully compatible with OPT (Open Pre-trained Transformer)-based variants. Experimental results on the COCO (Common Objects in Context) benchmark demonstrate consistent performance improvements, where ECMA improves the CIDEr (Consensus-based Image Description Evaluation) score from 144.6 to 146.8 and the BLEU-4 score from 42.5 to 43.9 on the OPT-6.7B model, corresponding to relative gains of 1.52% and 3.29%, respectively, while also achieving competitive METEOR (Metric for Evaluation of Translation with Explicit Ordering) scores. Further evaluations on social media datasets show that ECMA generates more coherent, context-aware, and socially informative captions, particularly for images involving complex interactions and socially meaningful scenes. Full article

Dual-target drug design addresses complex diseases and drug resistance, yet existing computational approaches struggle with simultaneous multi-protein optimization. This study presents SFG-Drug, a novel dual-target molecular generation model combining Monte Carlo tree search with gated recurrent unit neural networks for simultaneous MEK1 and mTOR targeting. The methodology employed DigFrag digital fragmentation on ZINC-250k dataset, integrated low-frequency masking techniques for enhanced diversity, and utilized molecular docking scores as reward functions. Comprehensive evaluation on MOSES benchmark demonstrated superior performance compared to state-of-the-art methods, achieving perfect validity (1.000), uniqueness (1.000), and novelty (1.000) scores with highest internal diversity indices (0.878 for IntDiv1, 0.860 for IntDiv2). Over 90% of generated molecules exhibited favorable binding affinity with both targets, showing optimal drug-like properties including QED values in [0.2, 0.7] range and high synthetic accessibility scores. Generated compounds demonstrated structural novelty with Tanimoto coefficients below 0.25 compared to known inhibitors while maintaining dual-target binding capability. The SFG-Drug model successfully bridges the gap between computational prediction and practical drug discovery, offering significant potential for developing new dual-target therapeutic agents and advancing AI-driven pharmaceutical research methodologies. Full article

Gadolinium-based contrast agents (GBCAs) are vital for glioma imaging yet pose safety, cost, and accessibility issues; predicting contrast enhancement from non-contrast MRI via machine learning (ML) provides a safer, economical alternative, as enhancement indicates tumor aggressiveness and informs treatment planning. However, scanner and population variability hinder robust model selection. To overcome this, a stability-aware framework was developed to identify reproducible ML pipelines for predicting glioma contrast enhancement across multicenter cohorts. A total of 1367 glioma cases from four TCIA datasets (UCSF-PDGM, UPENN-GB, BRATS-Africa, BRATS-TCGA-LGG) were analyzed, using non-contrast T1-weighted images as input and deriving enhancement status from paired post-contrast T1-weighted images; 108 IBSI-standardized radiomics features were extracted via PyRadiomics 3.1, then systematically combined with 48 dimensionality reduction algorithms and 25 classifiers into 1200 pipelines, evaluated through rotational validation (training on three datasets, external testing on the fourth, repeated across rotations) incorporating five-fold cross-validation and a composite score penalizing instability via standard deviation. Cross-validation accuracies spanned 0.91–0.96, with external testing yielding 0.87 (UCSF-PDGM), 0.98 (UPENN-GB), and 0.95 (BRATS-Africa), averaging ~0.93; F1, precision, and recall remained stable (0.87–0.96), while ROC-AUC varied (0.50–0.82) due to cohort heterogeneity, with the MI + ETr pipeline ranking highest for balanced accuracy and stability. This framework enables reliable, generalizable prediction of contrast enhancement from non-contrast glioma MRI, minimizing GBCA dependence and offering a scalable template for reproducible ML in neuro-oncology. Full article

This pilot study evaluated the visionMC system, a low-cost artificial intelligence system integrating HOG-based facial recognition and voice notifications, for workflow optimization in a family medicine practice. Implemented on a Raspberry Pi 4, the system was tested over two weeks with 50 patients. It achieved 85% recognition accuracy and an average detection time of 3.4 s. Compared with baseline, patient waiting times showed a substantial reduction in waiting time and administrative workload, and the administrative workload decreased by 5–7 min per patient. A satisfaction survey (N = 35) indicated high acceptance, with all scores above 4.5/5, particularly for usefulness and waiting time reduction. These results suggest that visionMC can improve efficiency and enhance patient experience with minimal financial and technical requirements. Larger multicenter studies are warranted to confirm scalability and generalizability. visionMC demonstrates that effective AI integration in small practices is feasible with minimal resources, supporting scalable digital health transformation. Beyond biometric identification, the system’s primary contribution is streamlining practice management by instantly displaying the arriving patient and enabling rapid chart preparation. Personalized greetings enhance patient experience, while email alerts on motion events provide a secondary security benefit. These combined effects drove the observed reductions in waiting and administrative times. Full article

Simulating damage phenomena in masonry structures remains a significant challenge because of the intricate and heterogeneous nature of this material. An accurate evaluation of fracture behavior is essential for assessing the bearing capacity of these structures, thereby mitigating dramatic failures. This paper proposes an innovative adaptive concurrent multiscale model for evaluating the bearing capacity of in-plane masonry structures under in-plane loadings. Developed within a Finite Element (FE) set, the proposed model employs a domain decomposition scheme to solve a combination of fine- and coarse-scale sub-models concurrently. In regions requiring less detail, the masonry is represented by homogeneous linear elastic macro-elements. The material properties for these macro-elements are derived through a first-order computational homogenization strategy. Conversely, in areas with higher resolution needs, the masonry is modeled by accurately depicting individual brick units and mortar joints. To capture strain localization effectively in these finer regions, a Phase Field Cohesive Zone Model (PF-CZM) formulation is employed as the fracture model. The adaptive nature derives from the fact that at the beginning of the analysis, the model is entirely composed of coarse regions. As nonlinear phenomena develop, these regions are progressively deactivated and replaced by finer regions. An activation criterion identifies damage-prone regions of the domain, thereby triggering the transition from macro to micro scales. The proposed model’s validity was assessed through multiscale numerical simulations applied to a targeted case study, with the results compared to those from a direct numerical simulation. The results confirm the effectiveness and accuracy of this innovative approach for analyzing masonry failure. Full article

The Cutting Development Chamber (CDC) design is presented as an innovative solution to crucial human challenges, such as food and plant medicinal production. Unlike conventional propagation chambers, the CDC is a much more comprehensive research tool, specifically designed to optimize plant reproduction from cuttings. It maintains precise control over humidity, temperature, and lighting, which are essential parameters for plant development, thus maximizing the success rate, even in difficult-to-propagate species. Its modular design is one of its main strengths, allowing users to adapt the chamber to their specific needs, whether for research studies or for larger-scale propagation. The most distinctive feature of this chamber is its ability to collect detailed, labeled data, such as images of plant growth and environmental parameters that can be used in artificial intelligence tasks, which differentiate it from chambers that are solely used for propagation. A study that validated and calibrated the chamber design using cuttings of various species demonstrated its effectiveness through descriptive statistics, confirming that CDC is a powerful tool for research and optimization of plant growth. In validation experiments (Aloysia citrodora and Stevia rebaudiana), the system generated 6579 labeled images and 67,919 environmental records, providing a robust dataset that confirmed stable control of temperature and humidity while documenting cutting development. Full article

In the era of Industry 5.0, the digital revolution stands as the paramount tool for achieving efficiency and elevating the standards of quality and quantity. This study delves deeply into the invaluable applications of digital twins within real production settings, highlighting their transformative potential across a multitude of industries. Focusing particularly on textiles, machinery, and electronics manufacturing, the authors illustrate how digital twins enhance productivity, anticipate challenges, bolster the food supply chain, refine healthcare services, and propel sustainability initiatives within each sector. Through concrete examples, we demonstrate how digital twins can markedly decrease waste, energy consumption, and production downtime, all while elevating product quality and enabling virtualization. By virtually simulating physical systems, numerous operational issues can be mitigated, underscoring the pivotal role of digital twins in fostering hyper-personalization, sustainability, and resilience the foundational tenets of Industry 5.0. Nevertheless, this evaluation acknowledges the inherent challenges associated with the widespread adoption of digital twins, including concerns regarding data infrastructure, cybersecurity, and workforce adaptation. By presenting a balanced assessment of both the advantages and disadvantages, this review aims to guide future research and development endeavors, paving the way for the successful integration of this revolutionary technology as we journey toward Industry 5.0. Full article

AI-based educational chatbots can expand access to learning, but many remain limited to text-only interfaces and fixed infrastructures, while purely generative responses raise concerns of reliability and consistency. In this context, we present ROboMC, a portable and multimodal system that combines a validated knowledge base with generative responses (OpenAI) and voice–text interaction, designed to enable both text and voice interaction, ensuring reliability and flexibility in diverse educational scenarios. The system, developed in Django, integrates two response pipelines: local search using normalized keywords and fuzzy matching in the LocalQuestion database, and fallback to the generative model GPT-3.5-Turbo (OpenAI, San Francisco, CA, USA) with a prompt adapted exclusively for Romanian and an explicit disclaimer. All interactions are logged in AutomaticQuestion for later analysis, supported by a semantic encoder (SentenceTransformer—paraphrase-multilingual-MiniLM-L12-v2’, Hugging Face Inc., New York, NY, USA) that ensures search tolerance to variations in phrasing. Voice interaction is managed through gTTS (Google LLC, Mountain View, CA, USA) with integrated audio playback, while portability is achieved through deployment on a Raspberry Pi 4B (Raspberry Pi Foundation, Cambridge, UK) with microphone, speaker, and battery power. Voice input is enabled through a cloud-based speech-to-text component (Google Web Speech API accessed via the Python SpeechRecognition library, (Anthony Zhang, open-source project, USA) using the Google Web Speech API (Google LLC, Mountain View, CA, USA; language = “ro-RO”)), allowing users to interact by speaking. Preliminary tests showed average latencies of 120–180 ms for validated responses on laptop and 250–350 ms on Raspberry Pi, respectively, 2.5–3.5 s on laptop and 4–6 s on Raspberry Pi for generative responses, timings considered acceptable for real educational scenarios. A small-scale usability study (N ≈ 35) indicated good acceptability (SUS ~80/100), with participants valuing the balance between validated and generative responses, the voice integration, and the hardware portability. Although system validation was carried out in the eHealth context, its architecture allows extension to any educational field: depending on the content introduced into the validated database, ROboMC can be adapted to medicine, engineering, social sciences, or other disciplines, relying on ChatGPT only when no clear match is found in the local base, making it a scalable and interdisciplinary solution. Full article