This article proposes a linear-time correlation-controlled shuffling method that attenuates Pearson correlation by permuting one variable through local swaps on a random regular graph. Each round processes a fixed edge order and accepts only vertex-disjoint swaps, producing greedy matching rather than a standard interchange-process update. The analysis uses the induced first-moment operator. A uniform lower bound on edge acceptance yields a Laplacian-domination relation, while a lower bound on the probability that vertices remain unmatched controls the negative spectrum. Conditional on the loop-deleted realized graph being connected and having a spectral gap bounded away from zero, the absolute value of the expected correlation contracts exponentially. For every fixed same-sign attenuation target, an activation probability exists that attains the target in expectation. Finite-sample calibration uses a fixed-budget multi-resolution direct search without assuming monotonicity. Because the procedure applies only permutations, the marginal distribution is preserved exactly. Under fixed degree and fixed calibration budgets, both calibration and shuffling scale linearly with sample size. Experiments show small target errors, exact marginal preservation, and approximately linear runtime scaling. A matched-degree spectral ablation further shows that, with degree and edge count held fixed, the higher-gap random regular graph exhibits substantially faster decay of the mean correlation than the low-gap regular circulant graph.
Industry 5.0 places autonomous agents inside its human–agent collaborative loop, resting its human-centricity pillar on an assumption of trustworthy collaboration that has not been examined critically. Agentic artificial intelligence has moved from research demonstration to industrial deployment within months, introducing a threat class this setting has not yet addressed. This article bridges three literatures developed in isolation, Industry 5.0 cybersecurity, agentic AI security, and Industry 5.0’s own foundational scholarship, proposing a six-category threat taxonomy and a paired forensic readiness framework, formalised as a five-level maturity model and grounded in real 2024–2026 incidents rather than hypothetical scenarios. Both contributions are evaluated through two complementary methods. An exploratory elicitation exercise, modelled on Delphi methodology and using six independent large language models as blind panellists across two rounds, converges on a specific structural critique that is incorporated into the taxonomy’s final design. A retrospective coding exercise then applies that taxonomy to fourteen publicly documented incidents, finding that half require multi-category classification and that two categories remain unexercised in the current public record, evidence that physically embodied industrial deployment has outpaced the documented incident base rather than a gap in the taxonomy itself. This urgency is reinforced by emerging EU and UK regulation imposing 24- and 72-h incident reporting obligations that, on current evidence, most industrial organisations are not positioned to meet. The article closes with a research agenda addressing liability, evidentiary standards, and the readiness-sustainability trade-off.
City information systems are steadily relying on various stakeholders, heterogeneous resources, and interrelated digital services. Nevertheless, many initiatives still face difficulties in aligning actors, business processes, data, and technological resources within a unified information systems architecture. Accordingly, this study presents and evaluates a federated enterprise architecture-based model for value co-creation in city information systems. The studied model investigates how strategy and capture, relationship management, and community context contribute to federated enterprise architecture, and how the architecture structuring supports service value co-creation, continuous improvement, and global impact. To validate the research model, data were gathered using a structured questionnaire that was administered to actors involved in information system initiatives. The studied hypotheses were tested using partial least squares structural equation modeling. The main findings demonstrate that community context has a positive influence on enterprise architecture structuring, which, in turn, supports service value co-creation through the alignment of stakeholders, processes, data, and digital services. Service value co-creation then contributes to continuous improvement and ecosystem global impact, acting as a mediating mechanism between information system to city service findings. Thus, the research contributes to the information systems literature by highlighting how federated architecture contributes positively to digital service in a multi-actor ecosystem and supports value co-creation in the city information system.
The growing volume of unstructured clinical and genetic text calls for automated processing methods that can meaningfully support clinical decision-making. Most existing work, however, treats the analysis of patient clinical descriptions and the interpretation of genetic reports as unrelated tasks, disregarding the sequence in which real diagnostic decisions unfold. We study two stage-specific families of classification tasks corresponding to the two points at which such decisions are taken—pre-genetic triage from clinical narratives, and post-genetic interpretation of completed genetic reports—and evaluate each family under a single leakage-controlled protocol using TF-IDF text representation with classical machine learning. All experiments use a corpus of 546 records with confirmed provenance, partitioned by a group-constrained master split (349 train/88 validation/109 test, seed 42) in which all 181 text-similarity candidate pairs are confined to a single partition. Task subsets contain 299 and 326 documents at the first stage and 375 and 461 at the second. Text representation relied on Word TF-IDF, Character TF-IDF and combined Word + Character TF-IDF with Logistic Regression and Linear SVM; every configuration was selected on validation Macro F1 alone, with the test partition opened only after selection. RuBERT-tiny2, XLM-RoBERTa base and PubMedBERT were fine-tuned as transformer baselines over three seeds each. Under this protocol, the first stage reached Macro F1 = 0.6876 (Accuracy = 0.8308, ROC-AUC = 0.8500) and the second-stage Macro F1 = 0.8225 (Accuracy = 0.8571, ROC-AUC = 0.9377) for the binary classification of diagnostic status. A controlled comparison holding the documents, split, vectorizer, classifier and hyperparameters fixed and varying only the label version increased Accuracy from 0.5738 to 0.8361 (exact McNemar p = 0.000145), whereas Macro F1 increased numerically from 0.5165 to 0.5966 but not significantly (paired bootstrap p = 0.354; 95% CI of the difference [−0.091, 0.229]). Two Russian and multilingual encoders collapsed to the majority class (Macro F1 = 0.4583), while an English biomedical encoder scored higher than the classical comparator (0.7705 ± 0.0119 against 0.6876) without reaching significance (paired bootstrap p = 0.463). Thus, the controlled experiment supports improved overall correctness, mainly associated with the majority class, but does not establish improved balanced class-wise performance at this sample size. In the second-stage leakage analysis, removing the diagnostic conclusion reduced Macro F1 from 0.8225 to 0.6478 (paired bootstrap p = 0.001), masking the exact label-generating rules reduced it to 0.7455 (p = 0.021), and combined masking reduced it to 0.6389 (p < 0.001). The unmasked Stage 2 score therefore depends materially on explicit report cues and should not be interpreted as evidence of independent diagnostic reasoning. The contribution of this work is a stage-oriented formulation of clinical genetic text classification in which the target variable of each task is aligned with the information available at the corresponding point of the diagnostic pathway, evaluated under a reproducible leakage-controlled protocol with bootstrap confidence intervals and paired significance testing. A unified single-model baseline on the joint target reached Macro F1 = 0.5238, below either stage-specific model on its own target. Because the labels are rule-derived rather than independently expert-validated, the corpus is small, and no external validation was performed, these results characterize what is achievable on this corpus rather than demonstrating clinical readiness.
The electrocardiogram (ECG) is a crucial non-invasive tool for diagnosing cardiovascular diseases. The standard 12-lead ECG configuration provides complementary spatio-temporal projections of cardiac electrical activity from distinct anatomical perspectives, enabling comprehensive pathological assessment. Unsupervised multi-lead ECG abnormality recognition has emerged as a promising approach, yet existing methods suffer from two limitations: insufficient attention to key pathological features and neglect of heterogeneous lead sensitivity. To address these issues, we propose the concept of lead-specific pathological sensitivity and develop the Lead-specific Pathological Sensitivity-aware Network (LPS-Net), comprising three modules: a Multi-scale Lead-specific Spatio-Temporal Encoder for discriminative representations, a Lead-specific Pathological Sensitivity-aware Topological Consensus Module that adaptively captures heterogeneous lead contributions, and an Asymmetric Hybrid Decoder that uses topological consensus to regularise reconstruction. Extensive experiments on four public 12-lead ECG benchmarks (PTB-XL, Chapman, CPSC-2018, and SPH) demonstrate the effectiveness of LPS-Net. On PTB-XL, LPS-Net achieves a clustering accuracy of 0.76 and a Macro F1 of 0.74 across five pathological superclass categories without any label information, outperforming the state-of-the-art method by 3%. Cross-dataset transfer experiments further confirm its generalisation ability. Ablation studies verify that the improvement originates specifically from the pathological-sensitivity formulation, and interpretability analyses reveal that the learned lead weight distributions align with clinical prior knowledge.
Guiding user attention across multiple spatial regions is essential for safety-critical activities such as driving and cycling. While peripheral visual cues can redirect attention without disrupting central vision, their effectiveness may diminish under high visual workload or when targets fall outside the field of view. Haptic feedback offers an alternative channel that operates independently of visual demands. This study systematically compares three feedback conditions in a multi-region target search task using a head-mounted display: Ambient Display, Haptic Display, and Ambient-Haptic Display. Results show that haptic feedback significantly outperformed peripheral visual cues for rear visual area search. While the multimodal combination did not further improve objective performance over haptic feedback alone, it lowered perceived mental and physical demand relative to peripheral visual cues and was rated highly by participants. These findings demonstrate that haptic feedback provides superior attention guidance for targets outside the current field of view. Our work offers practical guidance for designing cross-modal assistive systems in safety-critical environments.
Micro, small, and medium-sized enterprises (MSMEs) increasingly adopt digital technologies, yet converting this exposure into enterprise performance (ENP) requires multidimensional readiness capabilities. Evaluating these capabilities remains complex as existing studies often separate expert judgement, measurement validation, linear association, and explainable prediction. This study therefore develops an integrated analytical framework comprising two sequential phases. First, 22 experts contextualised the seven capability domains through a three-round Delphi process, followed by fuzzy TOPSIS to derive priorities under linguistic uncertainty. Second, cross-sectional survey data from a non-probability convenience sample of 610 eligible MSME respondents in Thailand underwent Confirmatory Factor Analysis (CFA) to validate the measurement structure, followed by hierarchical regression to establish a conventional baseline of adjusted linear associations. Within the second phase, eXtreme Gradient Boosting (XGBoost) and SHapley Additive exPlanations (SHAP) were subsequently applied for internal out-of-sample evaluation and model interpretation. The findings show that experts assigned the highest priorities to digital financial services and digital financial access. The XGBoost model achieved a test R2 of 0.6303, compared with 0.6282 for the OLS benchmark, indicating only a modest predictive improvement. Test-set SHAP identified social media use (mean |SHAP| = 0.2027), adaptive financial resilience, and financial management practice as the leading group of capability-level predictive contributors. The evidence is observational and does not support causal inference. This study contributes an integrated, measurement-validated, explainable framework for capability assessment. Expert-based priorities and SHAP-based predictive contributions provide complementary rather than equivalent forms of evidence.
The rapid growth of organisational textual data has increased the value of Natural Language Processing (NLP) and semantic analytics for strategic decision support. However, many employees still lack the knowledge and skills needed to evaluate AI-generated information critically. This study develops an explainable Management Information Systems (MIS) framework that integrates NLP-based semantic analytics with PLS-SEM to examine the relationship between AI illiteracy and strategic decision quality. A convergent mixed-methods design with sequential analytical integration was used with a valid sample of 200 knowledge workers from public organisations in Saudi Arabia across six industries. The sample included employees from Saudi Arabia, Egypt, Jordan, Sudan, Syria, India, and the Philippines. Quantitative data were analysed using PLS-SEM, while textual data were analysed using Sentence-BERT, BERTopic, semantic network analysis, and Aspect-Based Sentiment Analysis. The results showed that higher AI illiteracy was negatively associated with strategic decision quality and positively associated with automation bias, uncritical trust in AI, and cognitive offloading. Digital proficiency and AI governance awareness weakened the negative association between AI illiteracy and decision quality, while functional-background differences were examined through multigroup analysis. The semantic analysis identified six themes: AI competency, decision trust, AI governance, decision support, organisational learning, and risk awareness. Sentiment analysis showed positive views of productivity and decision support, together with concerns about algorithmic bias, explainability, transparency, and AI governance. The study contributes an integrated human–AI decision vulnerability framework in which semantic evidence complements structural modelling and provides a clearer understanding of AI-related competency, reliance, governance, and decision-support issues.
Point clouds are widely used to represent 3D visual content in immersive media, digital twins, and autonomous systems, but acquisition, compression, transmission, and rendering can introduce visible geometry and attribute distortions. Full-reference point cloud quality assessment (FR-PCQA) aims to predict the perceptual quality of a distorted point cloud by comparing it with a reference. A reliable FR-PCQA model should consider both the perception of 3D content by the human visual system via projected views and the manifestation of quality degradation in the point cloud geometry, color, and spatial structure. In this paper, we propose a multi-view projection and 3D feature fusion model for FR-PCQA. The proposed model integrates two complementary branches. In the projection branch, DISTS is applied to multi-view renderings aligned with the reference to capture perceptual similarity, and an additional six groups of geometric and photometric fidelity features (e.g., occupancy, depth fidelity and gradient domain fidelity) are developed to describe explicit geometric and photometric differences in the projected observations. In the 3D Feature Fusion branch, PCQM measures local geometry and color degradation, while a global structural descriptor with eight groups covering point count, position, scale, spatial distribution, and density is constructed to characterize the global properties of point clouds. Finally, a gradient boosting regression tree (GBRT) regressor is employed to predict the final quality score. Extensive experimental results show that the Spearman rank order correlation coefficient (SROCC) values are 0.91537, 0.9101, and 0.9778 on the SJTU-PCQA, WPC, and ICIP2020 datasets, respectively, outperforming the existing PCQA methods. These results indicate that the proposed multi-view projection and 3D feature fusion model provides an accurate and interpretable solution for FR-PCQA.
Operational risk management (ORM) in financial institutions has traditionally relied on quantitative loss event databases that capture what went wrong, but rarely why from a human factors perspective. This paper proposes a comprehensive Human Factors Framework for ORM that integrates the Human Factors Analysis and Classification System (HFACS) taxonomy with deep learning and large language models (LLMs) to extract, classify, and predict human-factor-driven operational risks from unstructured consumer complaint narratives. Using the U.S. Consumer Financial Protection Bureau (CFPB) Consumer Complaints Database (300,000 banking narratives), we (1) define and validate an eight-factor HFACS-Banking taxonomy through a Delphi study with five domain experts (Cohen’s κ = 0.86), (2) compare three extraction approaches—regex, supervised BERT, and Mistral-7B zero-shot LLM—achieving 0.81 average F1 with the LLM-distilled BERT relative to Mistral-generated reference labels, (3) propose a hybrid HF-BERT-BiLSTM-Attention architecture that fuses contextual text embeddings with structured features and HFACS factor probabilities for predicting a four-class company-response-based complaint severity proxy, and (4) provide explainability through SHAP feature attribution applied to the Random Forest baseline and attention-weight visualization of the proposed neural model. Under a temporal split comprising training data from 2014 to 2022, validation data from 2023H1, and held-out test data from 2023H2 to 2025, the proposed model achieves 91.42% accuracy and 89.78% macro F1 (95% CI from 1000-iteration paired bootstrap: [89.34, 90.21]), outperforming Random Forest (+9.84% F1, p < 0.001 Bonferroni-corrected), BiLSTM (+5.46%, p < 0.001), FinBERT (+3.21%, p = 0.003), and BERT-only (+3.92%, p = 0.002) baselines. Under the secondary random-split ablation analysis, removing the HFACS features and replacing additive attention with mean pooling reduced macro-F1 by 3.79 and 2.35 points, respectively. The findings demonstrate the retrospective feasibility of integrating theory-grounded human-factor representations with neural language models for complaint-outcome analysis; prospective institutional validation is required before operational use.
Long referring expressions create two coupled sources of hallucination in visual grounding. A detector can select an object that matches only part of the instruction, while a structured vision–language model (VLM) branch can hallucinate a target head or an attribute–object binding. We propose DeRecG, a training-free, anchor-preserved framework that addresses both sources. A frozen Grounding DINO detector supplies an anchor, and a frozen Qwen2-VL model decomposes the expression for candidate verification. Confidence-Constrained Disagreement Arbitration (CCDA) replaces the anchor only when a candidate passes head-validity, score-floor, score-margin, and spatial-disagreement gates. Lexical–Visual Candidate Expansion (LVCE) repairs parser-induced head errors before arbitration. Across six RefCOCO-family splits (46,842 expressions), LVCE+CCDA improves weighted accuracy at an intersection-over-union (IoU) threshold of 0.5 from 53.40% to 55.00%, while weighted mean IoU rises from 53.28% to 54.66%. With the same frozen thresholds, Ref-L4 validation accuracy rises from 31.02% to 34.73% and mean IoU from 33.43% to 36.72% over 13,420 long expressions; 646 anchor errors are corrected and 148 are induced (p<10−74), with no degradation on short-phrase Flickr30k Entities grounding. The complete pipeline requires no training, uses 6.10 GiB peak allocated GPU memory, and processes 1.15 images/s on an RTX 5090 D.
Automated cybergrooming detection can support the review of large volumes of online communication, but public datasets and detection approaches remain predominantly English-centered. This study presents a reproducible workflow for constructing target-language cybergrooming datasets and detection models from English resources, demonstrated for German. To the best of our knowledge, this is the first published study to construct and systematically evaluate German-language cybergrooming datasets and detection models. Four publicly available English cybergrooming datasets were converted into a shared structure and translated into informal German chat language. Translation quality was estimated using an aggregate of COMETKiwi and MetricX scores termed CometricX. Low-quality messages were retranslated and reevaluated. Cybergrooming messages were additionally augmented through German–Catalan backtranslation. The resulting datasets, with and without augmentation, were used to train and evaluate embedding-based linear support vector machines (SVMs) and ModernBERT models. Evaluation comprised translated test data and 156 manually labeled German conversations from five closed law-enforcement cases. The Perverted Justice Zig ChitChat dataset (PJZC) achieved the highest aggregate quality estimate and the most robust detection performance on real-world German conversations. In contrast, performance on machine-translated test data overestimated real-world detection performance. Furthermore, backtranslation yielded only limited and inconsistent improvements. Overall, the proposed workflow provides a practical starting point for developing target-language cybergrooming detectors while highlighting careful source dataset selection and evaluation on authentic target-language data.
Education systems are increasingly shaped by decisions about infrastructure, interoperability, data governance, procurement, and artificial intelligence, yet no established field treats these decisions as a single object of professional responsibility. This article asks whether Education Informatics can be specified as a distinct interface discipline, analogous to Business Informatics and Engineering Informatics. It is a critical conceptual review addressing four review questions; its search is documented using the PRISMA-ScR reporting items for transparency, without claiming scoping-review status. From Scopus and Web of Science, 1839 records were screened at title and abstract levels and analysed by framework synthesis against five adjacent fields. Four findings are reported, one per question. The term is in circulation but lacks a stable referent, definitional core, and citation community (RQ1). Procurement appears in under two per cent of records in every adjacent field, and only eleven of 785 education-relevant records integrate governance, architecture, and pedagogy (RQ2). The traditions able to ground the field are diagnostic rather than constructive (RQ3). No empirical study evaluates the institutional outcomes of interface-discipline formation (RQ4). Education Informatics is therefore advanced constitutively rather than descriptively: its object, theoretical foundations, competency profile, and three institutionalisation pathways are specified as claims to be tested.
The proliferation of rumors on social media has grown exponentially in recent years. Existing approaches focus on training models on single-domain datasets, resulting in limited generalization in cross-domain scenarios. Additionally, multi-domain rumor detection faces two critical challenges: bridging semantic gaps across heterogeneous domains and addressing modality-dependent dependencies across them. To address these issues, we propose a Multi-domain Aware Network (MDAN) specifically for multimodal rumor detection. MDAN synergizes adaptive domain embeddings with a domain self-augmentation mechanism to dynamically select expert knowledge while incorporating a multi-level knowledge fusion module that hierarchically integrates domain-shared and domain-specific representations via gated attention networks. These dual knowledge optimization strategies effectively address the aforementioned challenges, significantly enhancing model robustness in multi-domain detection scenarios. Extensive experiments on two public datasets demonstrate that MDAN outperforms state-of-the-art methods in accuracy and cross-domain generalization.
In precision agriculture, fruit segmentation serves as a fundamental visual prerequisite for orchard robotic operations, including precision management, automated harvesting and yield estimation. For unripe passion fruits, complex scene interference, high fruit–leaf similarity and foliage occlusion in practical orchard scenes easily cause missed detection and over-segmentation in existing models. To address these challenges, we build a multi-scene unripe-passion-fruit dataset named ZKMPF. Based on the UNet architecture, we propose Environment-Adaptive and Prompt-Fusion UNet (EAPF-UNet). First, EAPF-UNet embeds an Environmental Adapter into the encoder, which adjusts feature parameters to mitigate complex scenes interference. Then, it incorporates a Localization Multi-Scale Fusion Module (LMSM) and Refinement Multi-Scale Fusion Module (LMFM) to achieve accurate localization and refinement of unripe passion fruits and address scale variations. Finally, it designs the Multi-Dimensional Prompt Fusion module that integrates color, geometry and texture priors to improve the feature discriminability between unripe passion fruits and background. We conduct experiments on our self-built dataset, comparing EAPF-UNet with eight other segmentation models. Evaluated against eight segmentation models, EAPF-UNet obtains mDice of 85.51% and mIoU of 77.16% across six metrics and achieves competitive segmentation results within this dataset’s test scenes.
Churn prediction research is largely accuracy-oriented, and the link between prediction and financial decision-making remains underdeveloped. This study builds and tests the chain from prediction to profit as follows: calibrated prediction, SHAP-based explanation, a profit-maximizing threshold, and EMP-based evaluation, across five datasets in three sectors. Resampling and class weighting do not improve ranking quality (PR-AUC) and degrade calibration up to 19-fold (ECE); shifting the threshold recovers the same recall without retraining. SHAP-based analysis shows no driver is consistently strong across sectors: usage volume has the broadest reach, but the strongest drivers are dataset-specific, and models do not transfer after semantic alignment. The profit-maximizing threshold matches or beats the fixed 0.5 threshold in all 100 cost-success scenarios examined, and a paired test across datasets and seeds confirms the difference (Wilcoxon p < 0.001); a threshold that looks reasonable by accuracy can still cause a loss. Predictability and profitability rank inversely across datasets (Spearman ρ = −0.90): the most predictable dataset yields the lowest EMP, driven by churners’ value distribution. Compared with ProfLogit, which embeds the profit objective in training, a threshold on a well-calibrated model proves sufficient. Value comes from turning calibrated probabilities into decisions with a financial criterion, not from balancing data.
Reliable abnormal behavior recognition from surveillance videos is hindered by the high cost of clip-level annotation, the scarcity of abnormal samples, and the context-dependent nature of behavioral semantics. Although vision–language models offer strong semantic transferability, their application under limited supervision remains susceptible to noisy pseudo-labels and confirmation bias. We propose confidence-aware semi-supervised vision–language contrastive learning (CA-VLC), which jointly exploits limited labeled videos and abundant unlabeled videos. Building on an existing CLIP-initialized temporal backbone, CA-VLC combines behavior-only and context-enriched text prototypes through confidence- and agreement-guided semantic fusion. For unlabeled videos, the model generates predictions from weakly augmented views and selects reliable pseudo-labels using entropy-based confidence estimation and class-adaptive thresholds. Detached weak-view targets then supervise strongly augmented views through confidence-weighted self-training without requiring an additional teacher network. Furthermore, cross-view consistency regularization and confidence-aware contextual alignment suppress unreliable semantic cues and improve robustness to contextual noise. Experiments on CABR50 demonstrate consistent improvements across multiple labeled-data ratios, while evaluations on CABRZ6 and UCF-101 assess prompt-based transfer to predefined target label sets without target-domain fine-tuning. With 10% labeled videos, CA-VLC achieves 84.06% Top-1 accuracy and 83.51% Macro-F1, retaining 95.47% of its fully supervised Top-1 accuracy of 88.05%, thereby demonstrating its effectiveness for label-efficient abnormal behavior recognition.
Non-invasive blood pressure (BP) monitoring using photoplethysmography (PPG) has significant potential, yet accurately predicting systolic (SBP) and diastolic (DBP) blood pressure using photoplethysmogram (PPG) and electrocardiogram (ECG) signals remains challenging. This work proposes a novel dual-stream 1D encoder–decoder architecture for cuffless BP estimation from raw photoplethysmography (PPG) and electrocardiography (ECG) signals. The model incorporates multi-scale temporal feature extraction via dilated residual convolutions, cross-signal feature modulation at the bottleneck using Feature-wise Linear Modulation (FiLM), learned scale-wise fusion across encoder levels, attention-gated skip connections, and a hybrid mean-attention pooling regression head. Evaluated under calibration-based conditions on the PulseDB dataset (with selected data values of SBP ∈ [57, 180] mmHg and DBP ∈ [25, 100]), the proposed model achieved a mean absolute error (MAE) of 2.03 mmHg, mean error (ME) of 0.38 ± 3.13 mmHg, and an R2 of 0.93 for diastolic BP (DBP) and an MAE of 3.81 mmHg, ME of −0.28 ± 5.54 mmHg, and an R2 of 0.92 for systolic BP (SBP). The results meet the British Hypertension Society (BHS) and IEEE-1708 standard and achieved an “A” Grade. ECG alone provides lower prediction errors than PPG alone under the evaluated conditions, while their combination yields the best performance. Gender- and age-stratified analyses reveal consistent model behavior across demographic subgroups, with prediction error increasing modestly in older cohorts, particularly among women. This study provides an accurate, calibration-based, cuffless BP estimation and highlights its potential for non-invasive BP monitoring applications.
Enterprise Resource Planning (ERP) systems generate transactional data, but their value for process improvement depends on converting these records into reliable event logs and useful predictions. This study evaluated a framework combining event-log readiness assessment, process mining, predictive process monitoring, transparent machine learning, and domain validation in an Ecuadorian steel manufacturer. The analysis used 3740 production orders and 31,791 ERP-recorded events, split chronologically into training, validation, and out-of-time test samples. Process discovery identified execution heterogeneity; rework, accumulated waiting, route deviations, and resource congestion were positively associated with deadline violation. At production start, the process-aware ridge logistic model achieved a PR-AUC of 0.839, compared with 0.419 for the static representation and 0.683 for the remaining-slack benchmark, with an ROC-AUC of 0.893, a Brier score of 0.123, and a sensitivity of 0.804. The first material transaction was the earliest operationally useful checkpoint, preserving a median intervention window of 98.7 h before the committed completion date among predicted-positive cases. Additive model decomposition highlighted material-related waiting, queue waiting, partial material issues, and accumulated deviations as contributors to fitted risk. A nine-expert panel translated the evidence into six operational and tactical improvement opportunities. Findings support ERP-based early-warning process intelligence when readiness, temporal validation, calibration, transparency, and actionability are addressed jointly.
Poultry supply chains need to comply with Shariah requirements when supplying halal meat, which requires continuous quality improvement and multi-stakeholder inspection. However, current traceability systems have centralized opaque characteristics, fragmented records, and slow detection of anomalies, leading to food safety vulnerabilities and impractical recall protocols. To overcome these challenges, this paper presents an intelligent blockchain and Internet of Things (IoT)-based traceability system with a permissioned Hyperledger Fabric consortium network. A hybrid off-chain storage architecture supports scalable monitoring without ledger congestion: TimescaleDB stores high-frequency IoT sensor data (e.g., temperature and GPS), MinIO warehouses compliance documentation, while only immutable cryptographic hashes are stored on-chain to guarantee the integrity of the data. Halal governance is digitalized using role-based smart contracts that trigger real-time alerts, batch blocking, and automated recall upon environmental threshold breaches. Performance evaluation via Hyperledger Caliper indicates the system achieves 275 write transactions per second, a 300 TPS read throughput, an optimized read latency of 5 ms, and a write latency of less than 36 ms. Validated as a laboratory concept, this decentralized framework demonstrates proactive quality assurance, mitigates ledger bloat, and enhances halal-integrity trust.
