Search Results (3,506)

Remote fault-diagnosis services in dynamic edge–cloud environments depend on timely monitoring-data upload, remote inference, and result delivery, making their communication layer sensitive to variations in available bandwidth, link latency, and packet loss rate. This study addresses the network-state forecasting layer that supports such services rather than the fault-classification model itself. We propose Horizon-Aware Transformer–Ridge Fusion (HATR-Fusion), which combines a nonlinear Transformer expert with a low-variance ridge-regression expert for joint short- and long-horizon forecasting. Historical available bandwidth, link latency, packet loss rate, mobility, and offered load are used as inputs. Preprocessing statistics are estimated using the training set only, and validation-calibrated convex fusion weights are frozen before test inference. Experiments on controlled synthetic trajectories from eight links sampled at 1-min intervals, using 10 neural-network initialization seeds and ten baselines including DLinear and iTransformer, show that HATR-Fusion reduces mean absolute error (MAE) relative to the standalone Transformer by 6.94–8.31% over the 10-min horizon and by 3.02–4.40% over the 60-min horizon, with all six paired improvements remaining significant after Holm correction. Against iTransformer, HATR-Fusion is significantly more accurate for short-horizon bandwidth and latency, whereas iTransformer is significantly more accurate for long-horizon latency and packet loss; short-horizon packet loss and long-horizon bandwidth are not significantly different after Holm correction. The six-task mean normalized mean absolute error (NMAE) is 0.07106 for HATR-Fusion and 0.07047 for iTransformer, indicating comparable overall accuracy with task-dependent differences between the two methods. Ablation results show complementary short- and long-range contributions from ridge regression and Transformer, while input-quality sensitivity analysis identifies a limitation of the fixed fusion weights under corrupted or missing history. The conclusions are therefore restricted to scenarios with relatively stable input quality and distribution shifts comparable to those evaluated in this study. Full article

Multimodal Sequential recommendation alleviates the semantic insufficiency and data sparsity of item-ID-based models by incorporating side information such as text and images. However, multimodal systems face the dual challenges of feature-space heterogeneity and modality-specific noise, in addition to the dynamic evolution of user interests over time. Existing methods still struggle to jointly handle cross-modal alignment and time-aware preference modeling. To address these challenges, we propose a multimodal sequential recommendation framework with cross-modal alignment and temporal gating, which leverages item ID, text, and image modalities to capture users’ dynamic interests. The proposed model contains three core components. First, a cross-modal alignment mixture-of-experts module preserves modality-specific features with dedicated experts and captures shared semantics with common experts, thereby mitigating the semantic mismatch inherent in direct fusion. Second, a hierarchical time-aware mixture-of-experts module uses short-term intervals, long-term spans, and periodic time encodings for expert routing, and applies a time-aware modality gate to adaptively adjust the importance of ID, text, and image modalities under different temporal contexts. Third, a sequential interest contrastive learning objective enhances the discriminability of ID-based sequential interest representations by leveraging dynamic temperature scaling, multi-scale positive samples, hard negative mining, and diversity regularization. Experiments on games, beauty, and toys demonstrate that the proposed method consistently outperforms representative sequential and multimodal recommendation baselines on Normalized Discounted Cumulative Gain (NDCG)@5, NDCG@10, Mean Reciprocal Rank (MRR)@5, and MRR@10. Furthermore, ablation results validate the effectiveness of each proposed component. Full article

Nowadays, the volume of heterogeneous data in situational analysis centers is growing exponentially, leading to information overload for decision makers (DMs) and a decrease in the effectiveness of traditional decision support systems (DSS). Intelligent DSSs (ISDSS) demonstrate potential, but face challenges in explainability, heterogeneous data integration, cognitive load, and scalability. This paper proposes a method for intelligent decision support focused on identifying and generating options for resolving emergency situations—conditions that have no exact precedents in the knowledge base. The method includes formalizing the situation using a vector of normalized parameters $S\left(t\right)$, separating it into independent and dependent variables with the construction of a dependency tree, neural network classification of three types of conditions (normal, abnormal, and emergency) with a forecast for a lead interval τ, the synthesis of solutions for emergency situations based on an analysis of proximity graphs to known emergency precedents and evolutionary optimization. A computational experiment was conducted on the open dataset of the Tennessee Eastman Process simulation model with 28 failure types and 200 repeated simulations. The neural network classifier achieved an accuracy of 0.88 and a macro-averaged F1-score of 0.87 on a test set of 200 situations. Graphs of nearby emergency precedents were constructed for 50 synthetic emergency situations; analysis demonstrated the stability of topological characteristics (vertex degree 5.62 ± 1.18, closeness centrality 0.43 ± 0.09), substantiating the applicability of graph neural networks for accelerated control action synthesis. The proposed method reduces dependence on expert assessments and improves the adaptability and explainability of decisions, while the demonstrated stability of the graph-based precedent retrieval lays the groundwork for future full-scale validation of control-action synthesis in next-generation hybrid IDSS. Full article

Background/Objectives: Supervised electrocardiography-based sleep apnea detection often depends on dense segment-level annotations, creating substantial expert burden. To reduce this dependence, we investigated whether adjacent temporal context could improve classification when ground-truth apnea labels are limited. Methods: We systematically evaluated five context lengths and seven apnea-label retention levels across 35 experimental configurations using participant-grouped five-fold cross-validation with three training repetitions. Results: The results demonstrated that using multi-window temporal context consistently improved classification performance across the evaluated apnea-label retention range. All multi-window configurations achieved higher mean F1-scores than the corresponding single-window baseline, with the best-performing multi-window setting at each retention level providing gains of 7.18–12.23 percentage points. Context length showed a clear overall effect on performance, whereas no systematic interaction was observed between context length and apnea-label retention. Targeted repeated-center and random-shuffle control experiments further supported a contribution from temporally adjacent ECG information rather than sequence-length expansion, repeated target presentation, or temporally distant same-recording context. Conclusions: Overall, leveraging adjacent temporal context provides a practical strategy for improving ECG-based apnea classification and may be particularly useful for developing screening models when dense expert-provided apnea annotations are limited. Full article

Understanding complex vessel behaviors in port waters is important for maritime traffic supervision, navigation safety management, and intelligent maritime decision-making. However, existing AIS-based studies often focus on isolated behavior recognition or trajectory-level analysis, with limited integration of vessel attributes, navigation scenarios, motion states, and temporally organized behavioral processes. We develop a knowledge graph-driven framework for complex vessel behavior recognition and frequent behavior sequence pattern mining using AIS data. BehaviorEvents are constructed from continuous-navigation segments and integrated with water-area scenarios, motion states, vessel attributes, and temporal relationships to form a unified semantic representation. Based on this representation, interpretable semantic rules are used for event-level complex behavior recognition, while PrefixSpan is applied to Scene–SpeedState–TurningState token sequences to discover recurrent multi-event behavior patterns. Independent expert evaluation, semantic ablation, and sensitivity analyses are used to assess recognition credibility, contextual semantic constraints, and robustness, while a vessel-level Discovery–Validation strategy evaluates the reproducibility of frequent patterns. Experiments on AIS data from Xiamen Port waters involve 16,400 vessels, 239,877 continuous-navigation segments, and 2,457,965 BehaviorEvents, of which 624,561 match at least one predefined semantic rule or candidate condition. Independent expert evaluation of R1–R7 yields a macro-average confirmation rate of 92.11% and a Cohen’s κ of 0.746. Semantic ablation shows that Scene and VesselTypeClass provide important contextual constraints on broad motion-based rule activations, while sensitivity analyses indicate that the main recognition results remain stable under perturbations of motion-state, duration, and temporal-segmentation parameters. From 75,144 valid compressed behavior-token sequences, PrefixSpan identifies recurrent patterns involving medium-speed transit with course adjustments, low-speed–stop combinations, and maneuvering-related behaviors. The dominant Top-20 patterns showed substantial overlap and broadly consistent ranking across the vessel-level Discovery and Validation subsets, with a Jaccard overlap of 0.9048 and a Spearman rank correlation of 0.9654. Comparative evaluation with a normalized relational representation further shows equivalent analytical results, while the knowledge graph provides explicit organization of semantic relationships, temporal paths, and event-level traceability. These results indicate that the proposed framework provides a unified and interpretable semantic basis for connecting event-level complex vessel behavior recognition with sequence-level frequent behavior pattern mining in complex port environments. Full article

This study proposes an explainable artificial intelligence (XAI) engineering workflow for provider-level healthcare claim audit prioritization using SHAP-guided hybrid retrieval-augmented generation (RAG). The framework integrates provider-level claim aggregation, tree-based risk screening, SHAP explanation, exploratory group-level SHAP clustering, policy concept retrieval, and constrained large language model audit narrative generation. Experiments on a public Medicare provider fraud dataset use the dataset-provided PotentialFraud label as a weak audit prioritization label rather than a legal determination of fraud. The results show that reimbursement exposure, utilization duration, claim repetition, deductible patterns, and beneficiary case mix contribute to provider-level risk scores. Additional cross-validation, calibration, threshold, and scale-confounding analyses indicate that provider size and financial exposure are important confounders, while non-scale and contextual features also retain predictive information. Beyond prediction, the framework organizes local SHAP drivers into exploratory provider archetypes and maps explanation patterns to audit-relevant policy concepts. Compared with pure embedding retrieval, the SHAP-guided hybrid retriever increases policy concept diversity and explanation alignment, although these retrieval metrics do not replace independent expert audit validation. Because the public dataset does not include referral pathways, inter-facility transfers, provider–network relationships, or care-coordination records, the framework cannot determine whether utilization patterns are explained by clinically appropriate referrals, regional access constraints, or multi-level care pathways. Its current applicability is therefore limited to provider-level audit prioritization using the available claims and beneficiary variables. The proposed system is positioned as a reproducible engineering prototype for cautious, human-reviewed audit support rather than a comprehensive or automated fraud determination system. Full article

Although traditional machine learning models demonstrate high accuracy in agricultural prediction scenarios, their “black box” nature prevents them from transparently presenting decision-making mechanisms and limits their reliability in integrated pest management (IPM) processes. In this study, the Cherry Fruit Fly (Rhagoletis cerasi) was investigated. A rule-based, explainable artificial intelligence (XAI) framework is proposed for characterizing bio-edaphic profiles associated with observed Cherry Fruit Fly Pupal Count (CfPC) density levels and classifying microclimatic aspects (Aspects) using measurable edaphic and biological parameters. The developed hierarchical rule inference engine parses the decision trees of the LightGBM classifier, which achieved the highest performance when benchmarked against 10 baseline machine learning algorithms (11 models in total), and extracts human-interpretable results that can be directly interpreted by experts. In Experiment 1, the analysis characterized the combinations of observed CfPC and edaphic conditions associated with Low, Medium, and High pupal-density profiles, whereas Experiment 2 evaluated the classification of canopy aspect from the measured bio-edaphic variables. According to the derived rules, continuous biological counts (CfPC) serve as the primary biological reference, while edaphic parameters such as soil temperature, pH, lime content, and water saturation percentage characterize additional soil conditions associated with the observed pupal-density profiles. These synthesized rules provide an interpretable representation of the bio-edaphic patterns observed within the studied orchards and may support the development of future precision crop-protection strategies following independent validation. Full article

Accreditation and quality assurance in technical and vocational education and training (TVET) remain largely dependent on manual review and expert judgment. This study presents AccreditNet, an explainable tabular AI model embedded in a broader accreditation decision support framework. The experimentally evaluated component combines per-feature tokenization, multi-head self-attention, hierarchical criterion fusion, class-balanced focal loss, integrated gradients, and descriptive performance gap scoring. Evaluation is conducted against a structured NAAC benchmark that includes criterion scores and institutional attributes. Because the official NAAC grade is derived from criterion-level scoring, the benchmark is interpreted as a test of how well the model represents the criterion-to-grade mapping rather than as evidence of an independent predictive relationship. On the reported held-out split, AccreditNet achieved 93.72% accuracy, 92.84% macro F1-score, and 0.9706 AUC-ROC. OCR-based ingestion, NLP/LLM-assisted evidence mapping, automated reporting, and Saudi deployment are proposed operational extensions and were not evaluated in the present experiments. The results support further study of explainable accreditation analytics and decision support, subject to validation on local data. Full article

Accurate interwell connectivity characterization is critical for fine-grained waterflood optimization and reservoir management, often requiring integrated analysis across multiple disciplines and methods. Among these, injection–production response analysis stands as the most cost-effective and widely adopted approach. However, it remains highly subjective, heavily reliant on senior engineers’ decades of accumulated experience, and prohibitively labor-intensive for large-scale oilfields with hundreds of wells. With the exponential growth of production data in modern oilfields, manual analysis has become the bottleneck restricting the timeliness of reservoir management decisions. While signal processing techniques offer a promising path to automation, general-purpose algorithms fail to incorporate fundamental reservoir fluid flow laws, resulting in insufficient accuracy for practical engineering applications. To address this gap, we propose a novel connectivity identification method that mimics expert analysis logic by focusing on large-amplitude fluctuation segments rather than full-curve matching. Using curve slope as the core metric, cosine similarity quantifies trend consistency, while Root Mean Square Error (RMSE) measures amplitude proximity. Three targeted strategies enhance accuracy: key region screening with segmented matching, outlier removal accounting for time-varying lags, and multi-index weighted fusion. Validated on synthetic and mature carbonate waterflood field cases, the method improves the identification performance over benchmark Normalized Cross-Correlation (NCC) and Capacitance-Resistance Model (CRM) methods by more than 12% in both cases. It retains the reliability of traditional response analysis while achieving full automation, and can help estimate the timing of preferential flow path formation, requiring only routine production data to provide valuable reference for timely field development decision-making. Full article

Background: Integrating human factors into quantitative railway risk assessment remains methodologically unresolved. Failure Mode and Effects Analysis (FMEA) records human contributions as a single occurrence rating and cannot represent their organisational antecedents or their interactions. Purpose: This paper is a methodological proposal. It specifies the Integrated Human–Technical Risk Assessment (IHTRA) framework, which combines FMEA, the Human Factors Analysis and Classification System (HFACS) and Bayesian network modelling within the EN 50126 RAMS lifecycle, and demonstrates what such a specification makes analytically possible. It does not claim to validate the framework empirically. Methods: The Bayesian layer is specified in full as a ten-node network with all conditional probability tables reported. Of its twenty-two endogenous parameters, eight rest on evidence: four are derived from a survey of 89 Hungarian train drivers and four from the published fatigue literature. The remaining fourteen are declared structured assumptions awaiting expert elicitation. A conventional FMEA and the specified framework were both applied to national investigation report 2023-1152-5 (Sáp collision, 2023). Results: Human factors awareness yielded the highest domain mean (M = 4.23, SD = 1.18), with near-unanimous recognition of fatigue (M = 4.90) and workload (M = 4.87). Organisational consideration of human factors scored lowest (M = 2.66). No significant experience-group differences were observed (p > 0.05). The case analysis identified two HFACS levels as confirmed by the investigation findings and two further levels as plausible under the model interpretation. Inference over the specified network gives an illustrative, but not yet empirically calibrated, increase from p = 0.00038 to p = 0.00059 for the unsafe act and from p = 0.00011 to p = 0.00047 for the collision outcome. Conclusions: A specified but uncalibrated framework is a methodological contribution rather than an empirical one and is presented as such. This paper states precisely which parameters would have to be measured, and by what protocol, for the framework to become operational. Full article

Influence maximisation traditionally assumes that each activated neighbour contributes independently to the likelihood of a user adopting information, ignoring conjunctive synergies where a set of users must be active simultaneously to trigger another. Relational games provide a formal language for such coalitional dependencies, but the necessary influence hypergraph must be hand-crafted by domain experts, making them infeasible for large, dynamic social networks. We introduce the Neural Probabilistic Relational Game (N-PRG), a data-driven framework that automatically learns a probabilistic hypergraph of influence coalitions from cascade traces. A feed-forward neural network, trained via gradient descent to predict user activation, is interpreted using Deep SHAP to extract important set-level triggers. These are calibrated into a stochastic cascade model, the Probabilistic Relational Game (PRG), which generalises the Independent Cascade to set-based activation. We define the Minimal Reliable Seed Set problem, prove its NP-hardness even in the deterministic case, and establish that the expected influence function is monotone. We further demonstrate that, unlike the Independent Cascade model, the influence function under conjunctive (AND-type) hyperedges is in general not submodular, which precludes constant-factor approximation guarantees and motivates the use of greedy heuristics. Extensive experiments on synthetic data confirm that N-PRG successfully identifies coalitional interactions of size greater than one and achieves targeted out-of-sample coverage. Semi-synthetic experiments on Digg and Twitter network topologies demonstrate that N-PRG discovers seed sets up to 45% smaller than Independent Cascade baselines, while providing interpretable coalition pathways invisible to black-box methods. N-PRG thus unites the flexibility of gradient-descent learning with the structural rigour of relational games for influence analysis. Full article

Background: Patient experience has become a key component of healthcare quality assessment and improvement. Although its importance is increasingly recognized, there is a lack of standardized instruments specifically designed to assess patient experience in relation to nursing care during hospitalization. The starting point for this study is the ELIGES-Patient Experience questionnaire developed in 2024. This study aims to refine its content by incorporating the patient perspective and to evaluate its psychometric properties through a multicenter study. The final questionnaire will be intended to assess patient experience in relation to nursing care in inpatient units, providing a reliable tool for identifying opportunities for improvement and supporting person-centered care. Methods: A three-phase multicenter mixed-methods study will be conducted. First, a focus groups with expert patients will identify the key dimensions of patient experience related to nursing care during hospitalization. The initial questionnaire will then be reviewed and revised, yielding a first provisional version. Second, a national Delphi study involving expert patients will establish consensus on the questionnaire content, thereby yielding a consensus-based provisional version of the original instrument. Third, this provisional version of the questionnaire will undergo multicenter psychometric validation in six Spanish public hospitals. A pilot study will assess temporal stability and agreement, followed by administration of the provisional version to 2271 hospitalized patients to evaluate its psychometric properties. Expected Results: Following content refinement and psychometric validation, the resulting instrument is expected to provide a standardized measure of hospitalized patients’ experience of nursing care that may help identify areas for future evaluation and quality-improvement initiatives. Conclusions: Patient experience measurement lacks standardization, which hinders comparability and limits its use as a quality improvement tool. This highlights the need for instruments that are conceptually sound, methodologically robust, and operationally applicable. A distinctive feature of this protocol is the active involvement of patients throughout the entire process, in keeping with the principles of person-centered care and helping to strengthen the relationship between healthcare professionals and patients. Full article

Recent advances in Generative AI are creating new opportunities for engineering and geosciences activities. This study examined whether the initial production of Territorial and Environmental Diagnosis (TED) reports could be co-piloted by multi-agent systems. The study proposed a challenge between a classical engineering workflow and a multi-agent framework (MAF) combining specialized agents boosted by LLMs for data retrieval, Data Visualization, writing, statistical description, and publishing within a unified deterministic workflow. To evaluate the effectiveness of the MAF, a blind comparative assessment was conducted using sections of TED reports produced by internal experts for three French municipalities as reference case studies. The framework was tested across several key thematic sections, and MAF outputs were compared with the outputs generated by territorial experts. The evaluation focused on Data Relevance, Writing Quality, Data Visualization, Contextual Relevance, and Overall Rating. The results of our experiments indicate that MAF reduced, within tested conditions, the time required for reporting while maintaining a level of quality comparable to conventional reports. Evaluations confirmed that MAF performed well in themes relying on processing data, while human-produced reports retained an advantage in topics requiring deeper contextual interpretation. These findings support the potential of MAF as co-pilots for TED production in engineering workflows. Full article

Increasingly, residents prioritize diverse needs for residential buildings, hoping they can provide safe, comfortable, and healthy living environments. User satisfaction-based evaluation can significantly contribute to improving construction quality and optimizing construction management. Considering the strong subjectivity of existing assessment methods and the unreasonable allocation of indicator weights, this study establishes a feasible user satisfaction-based evaluation index system to provide a more objective evaluation of residential construction quality. By integrating the Analytic Hierarchy Process with the Fuzzy Comprehensive Evaluation method, a user satisfaction evaluation model is developed. A case study of a typical residential community in Zhongshan, Guangdong Province, was conducted to verify the scientific validity and applicability of the proposed model, demonstrating that it can effectively reduce the influence of subjective expert judgment. The results show that residents pay particular attention to leakage prevention and residential comfort, which is consistent with the current development trend of residential construction toward enhanced waterproofing performance and green development. The study provides real estate developers with theoretical support for improving construction quality management and optimizing residential living experience; contractors with practical references in optimizing the construction process; and urban–rural development departments with a scientific basis for strengthening supervision, creating acceptance standards, and implementing differentiated management policies. Full article

Reflectance Transformation Imaging (RTI) has long been employed across multiple research domains to document surface textures, inscriptions, and fine morphological features, including paleontology. However, RTI requires specialized equipment and controlled acquisition protocols that are often impractical in fieldwork or heterogeneous documentation contexts. To overcome these limitations, Virtual Reflectance Transformation Imaging (VRTI), based on high-resolution 3D models acquired through photogrammetry or structured-light scanning, has emerged as a promising alternative. This study investigates whether VRTI offers an operationally simpler yet perceptually consistent alternative to conventional RTI for dinosaur footprint interpretation, while providing a standardizàed FAIR-compliant VRTI workspace implemented in an open-source 3D environment. A comparative protocol was developed and validated through a classroom experiment involving paleobiology students, who performed footprint delineation and morphological interpretation using both RTI and VRTI datasets derived from a dinosaur footprint from the Geological Collection of the “Giovanni Capellini Museum” of the University of Bologna. The evaluation combines quantitative statistical analysis with qualitative expert assessment to compare interpretative performance, while optimal PTM-based raking-light conditions are also investigated. The results of the statistical comparison between RTI and VRTI remained non-significant (p > 0.05) for the tasks involved in the experiment, indicating comparable interpretative performance between the two approaches. Building on this validation, a preliminary usability assessment of the proposed workspace (n = 11) yielded a mean System Usability Scale (SUS) score of 67.7, supporting its practical applicability while providing a reproducible, interoperable, and scalable workflow for cultural heritage documentation. Full article