Search Results (57)

Equipment fault prediction in maglev train systems poses substantial challenges: fault events are inherently rare, class distributions are severely imbalanced, and individual equipment units are subject to complex spatial and functional couplings that single-device statistical approaches fundamentally cannot capture. To address these challenges, this study proposes a Knowledge Graph-driven Graph Neural Network (KG-GNN) framework. A fault knowledge graph encompassing equipment, fault, temporal, and environmental entities is constructed to unify multi-source maintenance data. Graph connectivity is established via three spatial relation types (co-location, co-zone, and co-level), with edge weights derived from Laplacian-smoothed Lift scores quantifying fault co-occurrence strength. A two-layer GATv2Conv-based graph attention network is designed: the first layer employs four-head attention with explicit edge-weight integration to capture heterogeneous neighborhood influences, while the second layer produces compact node embeddings via single-head attention. A Top-20 sparsification strategy suppresses weak-association noise, and training under severe class imbalance is stabilized through Focal Loss and F2-Score-guided early stopping. On the test set, the proposed method achieves an F2-Score of 0.5703, Recall of 0.6825, and AUC-ROC of 0.9329 (single-run evaluation); multi-seed evaluation (5 seeds) yields F2 = 0.5645 ± 0.0035, Recall = 0.6789 ± 0.0095, and AUC-ROC = 0.9298 ± 0.0026, outperforming the MLP baseline by 18.3% in F2-Score and substantially exceeding GCN (F2 = 0.1476 ± 0.0176) and GATConv (F2 = 0.4284 ± 0.0097). Ablation studies confirm the individual contributions of authentic graph topology, precise edge weighting, and graph sparsification to overall performance. Full article

Against the backdrop of the transition of new-type urbanization towards high-quality development, the triple contradictions of population agglomeration, land constraints, and housing supply-demand imbalance have become increasingly prominent. The conventional binary framework of human–land relations can no longer meet the requirements of coordinated development within human settlement systems, creating an urgent need to examine the multi-system interactions among population, land, and housing in order to resolve spatial mismatch. Taking Guangzhou as a case study, this research integrates 2020 population census data, land-use data from the European Space Agency (ESA), housing-price data from the Anjuke platform, and multi-source data on related influencing factors, and conducts a systematic empirical analysis by combining coupling coordination analysis, a relative development model, and the geographical detector. The findings reveal that the coupling coordination level of population, land and housing in Guangzhou exhibits a concentric, ring-shaped distribution pattern with central agglomeration and peripheral decline. The relative development among the three systems can be classified into matching types including the core-differentiated type, the peripheral-imbalanced type, and the surrounding-equilibrium type. With respect to influencing factors, all pairwise interactions are of the bi-factor enhancement type, and the driving mechanism displays a three-stage dynamic evolution. This study enriches research on human–land relations, provides precise guidance for optimizing spatial allocation and alleviating housing mismatch conflicts in Guangzhou, and offers transferable practical experience for comparable cities in China seeking to advance the high-quality development of new-type urbanization. Full article

Multi-hop question answering (MQA) requires models to perform multi-step reasoning and integrate multiple knowledge sources. However, existing methods combining pre-trained language models (PLMs) and graph neural networks (GNNs) often suffer from low computational efficiency, insufficient deep semantic fusion, and imbalanced modeling of heterogeneous relations. To solve these problems, we propose a Dynamic Hierarchical Adaptive Graph Convolution Network (DHACNet). First, to deal with the issues of insufficient computational efficiency and feature interpretability, we introduce Dynamic Sparse Activation (DSA). A trainable gate unit is used to generate importance masks for the encoder outputs, keeping only the task-relevant neurons. This greatly decreases the computational burden and enhances the interpretability of the model’s decisions. Second, to alleviate insufficient deep semantic fusion, we design a Hierarchical Feature Fusion (HFF) mechanism. It adaptively weights and fuses hidden states from different layers, enhancing the extraction and representation of deep textual semantics. Furthermore, for graph structure modeling, we present Adaptive Graph Convolution (AGC), which assigns learnable weights to different edge types in the graph, thereby improving heterogeneous relation modeling. Finally, hierarchical graph pooling is introduced, which integrates attention mechanism and Top-$K$ selection to achieve efficient and robust graph-level representation. The experimental results show that our proposed model maintains the symmetry between the text representation and graph representation through adaptive layered fusion and relational perceptual graph propagation. This symmetry-aware reasoning process encourages semantic consistency during multi-hop inference and makes knowledge integration more robust. Full article

The certification information of wheat varieties contains valuable breeding knowledge and plays an important role in germplasm resource management and breeding research. However, most certification information is stored in unstructured text form, making it difficult to support efficient knowledge acquisition and utilization. To address this issue, this paper investigates joint entity-relation extraction from wheat variety certification texts and its application to knowledge graph construction. Specifically, building on existing character-word fusion methods, we propose the Joint Entity-Relation Extraction for Wheat Variety Certification Texts (JERE-WVCT) to address indiscriminate incorporation of candidate features in character-word fusion that obscures key features, a lack of differentiated weight assignment for features, and severe imbalance across relation types. Within JERE-WVCT, a deep character-word fusion mechanism based on hierarchical filtering and ranking is designed to enhance the representation of domain-specific entities. In addition, relation labels are incorporated as prior knowledge to alleviate the impact of relation type imbalance and improve the model’s triple extraction capability. Experimental results show that JERE-WVCT achieves an F1 score of 96.78% on the wheat variety certification corpus, outperforming all baseline models. Based on the extracted triples, a wheat variety knowledge graph is constructed, and exploratory variety relationship analysis is conducted as a downstream application. The results demonstrate the effectiveness of the proposed model for structured knowledge acquisition and support graph-based exploration of wheat variety information. Full article

Knowledge graph link prediction is a fundamental task for improving the completeness and reasoning capability of knowledge graphs. In industrial knowledge graph scenarios, missing relations may limit knowledge completion, relational reasoning, and downstream industrial applications. Fault diagnosis is a representative application scenario, where missing relations among fault phenomena, alarm information, fault locations, and fault causes may further affect fault analysis, maintenance decision-making, and industrial knowledge services. Industrial knowledge graphs usually suffer from sparse local structures, imbalanced relation distributions, explicit entity-type boundaries, and highly confusing candidate entities with similar structural or semantic contexts. These characteristics make it difficult for conventional embedding-based or graph neural network-based methods to achieve reliable candidate ranking by relying only on structural propagation or semantic matching. To address these challenges, this study proposes a type-constrained structural–semantic fusion framework with dynamic relation priors for industrial knowledge graph link prediction, and further investigates its application to fault diagnosis. The proposed framework extends a relation-centered graph neural reasoning backbone by generating dynamic relation priors through query-conditioned relation-level graph propagation over a predefined relation graph, thereby enhancing query-specific structural reasoning. It further introduces a semantic projection module to align textual representations of entities and relations with structural representations at the candidate-ranking stage. In addition, relation-category and hierarchy-aware signals are used to modulate relation representations during propagation, while entity-type constraints are incorporated into final scoring and type-constrained hard negative construction. In this way, structural evidence, textual semantic information, and entity-type validity constraints are jointly used for candidate ranking rather than being treated as isolated signals. Experiments are conducted on two public benchmark datasets, WN18RR and FB15k-237, and two industrial knowledge graph datasets in Chinese and English. The Chinese industrial knowledge graph is constructed from fault diagnosis knowledge and is used as a representative application dataset, while the English industrial knowledge graph is used to further evaluate the adaptability of the proposed framework in a related industrial production scenario. The proposed method achieves MRR scores of 0.599 and 0.446 on WN18RR and FB15k-237, respectively, and obtains MRR scores of 0.8532 and 0.7994 on the Chinese and English industrial knowledge graphs. The results demonstrate that the proposed framework improves both general link prediction performance and industrial-domain adaptability, especially in scenarios involving sparse structures, type-constrained candidate validity, and semantically confusing entities, and shows practical potential for fault diagnosis applications. Full article

Fraud detection on multi-relational graphs is challenging because real-world fraud-related data often contains heterogeneous relations, noisy structures, and imbalanced labels. Existing GNN-based methods usually rely on predefined graph structures, but real-world financial graphs often contain noisy, redundant, or missing relations, which undermine neighborhood aggregation and message passing. In addition, single-view learning is insufficient to capture both local structural patterns and high-order semantic dependencies, limiting performance in complex fraud scenarios. To address this issue, we propose HGSO-DVCL, a heterogeneous graph structure optimization framework with dual-view contrastive learning. The framework performs type-aware structure optimization for each relation subgraph and integrates optimized graphs with the original structure through channel attention. A dual-view encoder then learns complementary representations from the network schema view and the meta-path view, while contrastive learning promotes consistency and complementarity between the two perspectives. An end-to-end objective jointly optimizes fraud classification, structure regularization, and contrastive alignment. Experiments on public multi-relational fraud detection benchmarks show that HGSO-DVCL achieves strong and competitive performance, while ablation and sensitivity studies support the effectiveness and stability of the proposed framework under the evaluated benchmark settings. Full article

The relationship between humans and the land has always been a topic in geographical studies. Northeast China, one of the regions with the shortest history in China, is also one of the regions most representative of changes in human–land relationships. However, scholars have rarely conducted quantitative region-scale research on the dynamic changes in, and drivers of, human–land relationships in this region. This study utilizes Landsat remote sensing imagery to identify changes in the distribution of land use types in Northeast China from 1990 to 2022. By constructing a human–land coordination model, it measures the intensity of human activity and levels of human–land coordination, analyzes their spatiotemporal dynamic characteristics, and further uses the Geodetector model to explore the factors driving and interactions influencing this evolution. (1) The results show that, from 1990 to 2022, the level of human–land coordination in Northeast China generally exhibited a spatial distribution pattern decreasing from northwest to southeast. The area of imbalanced human–land relationships continuously decreased, while coordinated areas steadily increased, indicating gradual improvement in human–land relations. The predominant type of coordination was moderate imbalance, with high imbalance as a secondary level. (2) The results also demonstrate that population size, GDP, and tertiary industry output have significant explanatory power regarding levels of human–land coordination. The importance of economic development level, natural resource endowment, and natural environmental characteristics to the evolution of human–land has progressively increased. Full article

Ramadan fasting substantially alters meal timing, sleep patterns, and daily activity, thereby increasing the risk of hypoglycaemia in adults with type 1 diabetes (T1D). Although continuous glucose monitoring (CGM) systems provide real-time alerts, these are largely reactive or limited to short prediction horizons, offering insufficient warning under fasting-related behavioural and circadian disruption. This study aims to evaluate whether behaviour-aware, temporally enriched recurrent deep learning models, leveraging multimodal CGM and wearable-derived signals, can forecast hypoglycaemia one hour ahead during Ramadan and the post-fasting period. In an observational, free-living cohort study conducted in Qatar, 33 adults with T1D were monitored using CGM and a wrist-worn wearable during Ramadan 2023 and the subsequent month. Multimodal data were aggregated into hourly features and organised into rolling 36 h sequences. In addition to physiological signals, explicit temporal and circadian proxy features were engineered, including cyclic time encodings, day–night indicators, and Ramadan-specific behavioural windows (e.g., pre-iftar, iftar, post-iftar, and fasting phases). Recurrent models, including LSTM and BiLSTM architectures, were trained using patient-wise, leak-free splits, with focal loss applied to address class imbalance. Model performance was evaluated on a held-out, naturally imbalanced test set using ROC AUC, precision–recall AUC, recall, and probability calibration, alongside cross-phase evaluation between Ramadan and post-fasting periods. Following quality control, 1164 participant-days were retained, with hypoglycaemia accounting for approximately 4% of hourly observations. Temporal feature enrichment and the use of a 36 h lookback window improved both discrimination and calibration, with performance stabilizing beyond this horizon. On the imbalanced test set, the best-performing multimodal model achieved an ROC AUC of 0.867 and a precision–recall AUC of 0.341, identifying 77% of next-hour hypoglycaemic events at a sensitivity-focused operating point (precision = 0.14). The selected BiLSTM model demonstrated good probability calibration (Brier score ≈ 0.03). Models trained using wearable-derived inputs alone achieved comparable discrimination and, in some configurations, higher precision–recall AUC than CGM-only baselines. Notably, models trained on the original imbalanced data outperformed resampled variants, suggesting that temporal and behavioural features provided sufficient discriminatory signal without requiring aggressive class balancing. Cross-phase evaluation indicated robust generalisation, particularly for the BiLSTM model. Overall, behaviour-aware, temporally enriched multimodal models can provide calibrated, hour-ahead hypoglycaemia risk estimates during Ramadan fasting in adults with T1D, enabling proactive intervention beyond reactive CGM alerts. Explicit modelling of circadian and behavioural dynamics enhances predictive performance under real-world class imbalance. Furthermore, integrating wearable-derived behavioural and physiological signals adds predictive value beyond CGM alone, supporting robustness across varying levels of contextual data availability. External validation and prospective clinical evaluation are required prior to deployment. Full article

Fault diagnosis in modern power systems increasingly depends on unstructured operation and maintenance (O&M) logs, yet real-world logs are often small in scale and highly imbalanced across fault types, which degrades the generalizability of standard neural models. This paper proposes a graph-augmented diagnostic framework that integrates imbalance-aware knowledge extraction with interpretable reasoning. The framework consists of three stages: (1) domain adaptation of a BERT–BiLSTM–CRF NER model and a BERT–MLP RE model using an imbalance-aware training recipe that combines Low-Rank Adaptation (LoRA), a mixed focal–range loss, and undersampling; (2) construction of a power-system knowledge graph that organizes extracted entities and relations (e.g., fault devices, abnormal phenomena, causes, and handling measures); and (3) a graph-augmented assistant agent that reuses the NER model as a graph-aware retriever within a retrieval-augmented generation (RAG) architecture to support contextualized and interpretable diagnostic reasoning. Experiments on 3921 real-world fault-processing logs show consistent gains: NER reaches 92.0% accuracy and 71.3% Macro-F1 (vs. 80.3% and 63.2%), and RE achieves 88.0% accuracy and 70.1% F1 (vs. 82.1% and 60.4%), while reducing average training time per epoch by about 18%. These results demonstrate an efficient and practical path toward robust log-based fault diagnosis under scarce and imbalanced data. Full article

Validated biomarkers for clinical decision-making in spondyloarthritis (SpA) remain limited, and exploratory experimental studies may help prioritize candidate immune and bone-related readouts for future validation. In this pilot study, cytokine and bone-related biomarker profiles were analyzed in a proteoglycan-induced SpA model using Th1-prone C57BL/6J wild-type (WT) mice (non-immunized n = 8; immunized n = 16) and Th2-prone BALB/c WT mice (non-immunized n = 7; immunized n = 9), as well as immunized TLR2-knockout (KO) (n = 7), TLR3-KO (n = 8), and TLR4-KO (n = 3) strains on the C57BL/6J background. Serum cytokines were quantified longitudinally with a 26-plex immunoassay, and ELISA measured bone metabolism markers (DKK1, Wnt3a, Noggin). Cytokine analysis revealed distinct Th1/Th2 polarization: immunized Th1-prone C57BL/6J WT mice exhibited high Th1- and Th17-type cytokines (TNF-α, IFNγ, IL-12p70, IL-17A, and IL-22), whereas immunized Th2-prone BALB/c WT mice showed elevated Th2- and eosinophil-related cytokines (IL-4, IL-9, IL-13, IL-5, and RANTES). In TLR2-KO and TLR3-KO, Th1- and Th17-associated cytokines were markedly reduced, while Th2 cytokines were increased, confirming that TLR2 is essential for maintaining pro-inflammatory signaling. DKK-1 and Noggin levels were significantly higher in TLR2-KO mice, indicating altered terminal serum bone-marker profiles under immunized conditions. These findings indicate that Th1/Th2 immune backgrounds and TLR-associated contexts are associated with distinct cytokine patterns and differences in terminal bone markers in this experimental SpA model. Given the pilot design, small and imbalanced groups, missing non-immunized TLR-KO controls, and exploratory statistics without multiplicity adjustment, the results should be interpreted as hypothesis-generating and require confirmation in appropriately controlled, statistically powered studies incorporating longitudinal and structural endpoints, as the present findings are exploratory and not directly translatable to clinical biomarker use or therapeutic decision-making. Full article

Software maintenance and release management demand proper identification of bug dependencies since priority violations or unresolved dependent issues can often lead to a chain of failures. However, dependency annotations in bug reports are extremely sparse and imbalanced. These dependencies are often expressed implicitly through natural language descriptions rather than explicit metadata. This creates challenges for automated multilabel dependency classification systems. To tackle these drawbacks, we introduce a meta-contrastive optimization framework (MCOF). This framework integrates established learning paradigms to enhance transformer-based classification through two key mechanisms: (1) a meta-contrastive objective adapted for enhancing discriminative representation learning under few-shot supervision, particularly for rare dependency types, and (2) dependency-aware Laplacian regularization that captures relational structures among different dependency types, reducing confusion between semantically related labels. Experimental evaluation on a real-world dataset demonstrates that MCOF achieves significant improvement over strong baselines, including BM25-based clustering and standard BERT fine-tuning. The framework attains a micro-F1 score of 0.76 and macro-F1 score of 0.58, while reducing hamming loss to 0.14. Label-wise analysis shows significant performance gain on low-frequency dependency types, with improvements of up to 16% in F1 score. Runtime analysis exhibits only modest inference overhead at 15%, confirming that MCOF remains practical for deployment in CI/AT pipelines. These results demonstrate that integrating meta-contrastive learning and structural regularization is an effective approach for robust bug dependency discovery. The framework provides both practical and accurate solutions for supporting real-world software engineering workflows. Full article

With the rapid development of artificial-intelligence technologies in the field of automated driving, many jurisdictions have successively adopted legislation and policy guidance to regulate the safe use of such technologies and to promote their orderly development. This article takes as its objects of study a set of jurisdictions that are particularly representative in terms of legislation and practice across different legal systems. The study finds that liability regimes for traffic accidents involving automated driving fall mainly into four types: the driver liability regime, the system liability regime, the manufacturer or operator liability regime, and the composite liability regime. In application, each of these regimes reveals different types of institutional dilemmas, including blurred boundaries of liability, underdeveloped mechanisms for evidence production and fact-finding, imbalanced allocation of liability, and fragmentation of the rules governing liability determination. In response to these dilemmas, this article proposes corresponding optimisation pathways, including clarifying the boundaries of driver liability and improving supplementary liability mechanisms; specifying in greater detail the obligations of system providers and strengthening data-related fact-finding rules; developing a reasonable allocation of liability between manufacturers and operators together with supporting insurance arrangements; and enhancing institutional coordination under the composite liability regime. These optimisation pathways not only provide institutional reference for jurisdictions seeking to maintain risk controllability while fostering innovation amid rapid technological evolution, but also lay the groundwork for the systematic improvement of future governance of automated driving. Full article

Due to limited land resources and traditional farming practices, continuous cassava cropping is common in China. This practice leads to soil degradation, including reduced fertility, imbalanced microbial communities, and lower crop yields. In this study, we investigated the impacts of continuous cassava cropping (CC) and cassava–maize rotation (RC) systems on soil physicochemical properties, microbial community composition, and functional gene abundance related to carbon and nitrogen cycling. The RC system consists of a five-year rotation cycle: cassava is planted in the first year, followed by two consecutive years of maize, and then, cassava is planted again in the last two years. The soil type is classified as Haplic Acrisols with a clay loam texture in this research. Soil samples from both cropping systems were analyzed for physicochemical properties and enzyme activities, and the results showed significant decreases in soil pH, available nitrogen, available phosphorus, and available potassium in CC. Using metagenomic sequencing, 1,280,928 and 1,224,958 unigenes were identified under RC and CC, respectively, with differences in microbial taxonomic and functional profiles. Bacteria accounted for 89.257% of the soil community in RC, whereas the proportion was 88.72% in CC. The proportions of eukaryota and viruses in RC were 0.031% and 0.006%, respectively; in contrast, their proportions were 0.04% and 0.02% in CC, respectively. Cassava–maize rotation promoted the metabolic activities of soil microbes, leading to a significant enhancement in functional genes related to nitrogen and carbon cycling, such as nasA, nasD, nrtC, coxA, porA, and frdA. This shows that microbial activity and nutrient cycling improved in the crop rotation system. Thus, these findings highlight the importance of crop rotation for maintaining soil health, enhancing microbial functions, and improving sustainable cassava production. This study provides valuable insights into the management of cassava agroecosystems and the mitigation of the adverse effects of continuous cropping. Full article

Ocular disease (OD) represents a complex medical condition affecting humans. OD diagnosis is a challenging process in the current medical system, and blindness may occur if the disease is not detected at its initial phase. Recent studies showed significant outcomes in the identification of OD using deep learning (DL) models. Thus, this work aims to develop a multi-classification DL-based model for the classification of seven ODs, including normal (NOR), age-related macular degeneration (AMD), diabetic retinopathy (DR), glaucoma (GLU), maculopathy (MAC), non-proliferative diabetic retinopathy (NPDR), and proliferative diabetic retinopathy (PDR), using color fundus images (CFIs). This work proposes a custom model named the ocular disease detection model (ODDM) based on a CNN. The proposed ODDM is trained and tested on a publicly available ocular disease dataset (ODD). Additionally, the SMOTE Tomek (SM-TOM) approach is also used to handle the imbalanced distribution of the OD images in the ODD. The performance of the ODDM is compared with seven baseline models, including DenseNet-201 (R₁), EfficientNet-B0 (R₂), Inception-V3 (R₃), MobileNet (R₄), Vgg-16 (R₅), Vgg-19 (R₆), and ResNet-50 (R₇). The proposed ODDM obtained a 98.94% AUC, along with 97.19% accuracy, a recall of 88.74%, a precision of 95.23%, and an F1-score of 88.31% in classifying the seven different types of OD. Furthermore, ANOVA and Tukey HSD (Honestly Significant Difference) post hoc tests are also applied to represent the statistical significance of the proposed ODDM. Thus, this study concludes that the results of the proposed ODDM are superior to those of baseline models and state-of-the-art models. Full article

Background/Objectives: Misinformation significantly impacts self-care behaviors and treatment outcomes in patients with type 2 diabetes mellitus (T2DM). We investigated the prevalence and content of diabetes-related misinformation among Thai patients with T2DM, examining the influence on cognitive attitudes and intentions to share such information. Methods: We employed a mixed-methods approach, conducting initial qualitative interviews with healthcare professionals and patients with T2DM to identify key misinformation themes. These themes guided the development of a validated questionnaire that was distributed to 107 patients with T2DM. Spearman’s correlation and multiple linear regression analyses were used to assess the relationships between misinformation exposure, attitudes, and sharing intentions. Results: Misinformation was categorized into four domains: medication side effects, alternative treatments, imbalanced lifestyle, and symptom perception. Exposure to misinformation ranged from 19.6% to 94.4%, with word of mouth identified as the primary source (81.18%). Misconceptions regarding symptom perception and alternative treatments were most prevalent. Information source, especially healthcare providers (β = 0.4); personal attitudes towards misinformation (β = 0.24); and exposure level (β = 0.46) significantly influenced the intention to share misinformation. Conclusions: This study highlights the need for targeted educational interventions to address widespread misconceptions in the management of T2DM, particularly those related to symptom perception and alternative treatments. Addressing these misinformation sources may be associated with improved self-management practices and could inform strategies aimed at enhancing patient outcomes. Full article