Search Results (209)

Temporal knowledge graph completion (TKGC) infers missing facts by modeling the temporal evolution of relations. Existing methods typically encode time through low-dimensional geometric transformations or frequency-domain decomposition. However, periodically recurring relations can still be mapped to overlapping trajectories in the same two-dimensional working space, which may reduce temporal separability and produce conflicting optimization signals. To address this limitation, we propose the Frequency-Modulated Spiral Manifold (FMSM) model, which includes the following: (1) a relation-adaptive spectral partition and relation-dimension gate for fusing long- and short-term relation channels; (2) an independent global phase embedding and nonlinear spiral push that lift entangled planar trajectories onto separated three-dimensional manifold layers; and (3) a spiral norm regularizer that stabilizes temporal evolution while preserving valid burst signals. The artificial-intelligence contribution of FMSM is a phase-conditioned spectral-geometric representation that separates recurring temporal facts while retaining multi-scale relation dynamics. Its engineering application is the completion of time-stamped event records for dynamic knowledge-based decision-support systems. Experiments on ICEWS14, ICEWS05-15, and GDELT show that, compared with the strongest reported baseline TeRDy, FMSM yields relative MRR improvements of 0.62%, 0.86%, and 13.28%. Full article

Low Earth orbit (LEO) mega-constellations demand scalable routing that survives time-varying topologies, constrained onboard resources, and dynamic traffic. Deterministic shortest-path routing guarantees optimal paths but adapts poorly to real-time loads; distributed deep reinforcement learning (DRL) can introduce loops and inconsistent end-to-end decisions. This paper fuses deterministic inter-domain planning with DRL-based intra-domain forwarding in a single hierarchical framework. An evolutionary greedy algorithm partitions the constellation into compact domains. Dijkstra’s algorithm then computes backbone paths on the domain-level graph. Inside each domain, a context-enhanced Per-Action Dueling Double Deep Q-Network encodes individual neighbors through a weight-shared encoder and summarizes the valid-neighbor set via masked mean pooling. This design lets the policy compare a candidate against the current alternative set without injecting input-order bias. Local one- and two-hop topological features drive decentralized inference. A greedy–beam–Dijkstra fallback ladder guarantees reachability whenever the subgraph stays connected. On a 1584-satellite Starlink Gen1-1 topology, all 21 domain sizes and six inter-domain strategies reach 100% of test pairs; the best average hop count sits at 1.16× the global Dijkstra benchmark. Under an identical 52-dimensional state and training pipeline on 1000 held-out source–destination pairs, Context Per-Action uses 75.8% fewer parameters than a flat multilayer perceptron (MLP), lifts greedy success from 74.6% to 83.5%, and lifts greedy-plus-beam success from 88.3% to 94.5% (means over three independent training seeds). Centralized load-aware routing under dynamic traffic cuts high-load packet loss from 34–73% to 0–9.5% in the adopted flow-level model and preserves 99.2% reachability despite 30% link failures. Zero-shot transfer from ideal Walker topologies to real two-line element (TLE) snapshots and purely local load adaptation remain open; multi-snapshot training or online adaptation is the necessary next step. Full article

Cross-condition fault identification for crane slewing bearings is difficult because low rotational speed, heavy loading, and operating-condition variation jointly weaken fault-induced impulses and alter the distribution of monitoring data. In addition, vibration and acoustic emission signals describe different aspects of bearing degradation, making fixed multi-sensor fusion insufficient when sensor sensitivity changes across fault states. This study proposes a Lightweight Domain-Adaptive Graph Convolutional Network (LDAGCN) for cross-condition diagnosis. The model uses six vibration channels and one acoustic emission channel as synchronized heterogeneous inputs. Two compact one-dimensional encoders first learn modality-specific temporal representations, after which a feature-wise gate determines the relative contribution of each sensing modality. The fused embeddings in every mini-batch are regarded as graph nodes, and a sparse sample graph is reconstructed from Top-k cosine similarities. Multi-receptive-field graph convolution then aggregates one-hop and higher-order neighborhood information, while a residual connection limits excessive modification of the original fused features. To reduce the discrepancy between operating conditions, the training objective combines source-domain classification, adversarial domain discrimination, maximum mean discrepancy, and supervision from a small labeled target-domain adaptation subset. Experiments were carried out on a dedicated crane slewing-bearing test rig containing normal, inner-race fault, outer-race fault, and B1 localized-fault states. On the target-condition test set, LDAGCN achieved an accuracy of 97.22%, a Macro-F1 score of 97.21%, a Macro-Precision of 97.37%, and a Macro-Recall of 97.22%. The proposed model also outperformed 1D-CNN, ResNet1D, CNN-LSTM, DANN, MMD-DAN, and DeepCORAL under the same data partition. The confusion matrix and t-SNE projection indicate that the learned representation reduces the source–target distribution gap while retaining fault-class separation. The ablation results further indicate that the acoustic emission information, gated fusion, graph-based association learning, and domain adaptation complement each other in terms of the final diagnostic performance, while maintaining a lightweight architecture that is suitable for practical monitoring. Full article

Automatic modulation recognition (AMR) is essential for cognitive radio and intelligent wireless communication, yet its performance degrades markedly under low signal-to-noise ratio (SNR) conditions because weak local structures are corrupted and class boundaries become less separable. To address this problem, we propose ADGNet, an automatic modulation recognition network based on adaptive wavelet enhancement and dynamic graph construction. ADGNet adopts a dual-branch architecture in which the main branch extracts time-domain amplitude–phase features from raw in-phase/quadrature sequences, while the auxiliary branch combines short-time Fourier transform features with adaptive wavelet features to capture complementary frequency–energy distributions and multi-scale transient details. A dual-domain adaptive encoder then recalibrates and fuses the two branches, suppressing redundant and noise-contaminated responses. In addition, a G2 dynamic topology module constructs sample-adaptive top-k adjacency matrices from node features and fuses them with the original graph structure to improve temporal relation modeling. Experiments on RadioML2016.10a and RadioML2016.10b show average accuracies of 64.23% and 69.69%, respectively. On RadioML2016.10b, ADGNet achieves 63.48% average accuracy from −12 dB to 0 dB, compared with 59.51% for the baseline. With approximately 0.13 million parameters, ADGNet provides a favorable balance between low-SNR robustness and model complexity. Full article

Adaptive learning systems need to sequence learning activities according to individual progress while respecting the prerequisite structure of the learning domain. This study proposes the Triadic Educational Knowledge Space (TEKS) framework for target-oriented adaptive sequencing in structured domains. TEKS integrates a prerequisite graph, a task-specific target set, and a recursively updated predictive distribution over prerequisite-consistent latent knowledge states. At each decision step, candidate items are evaluated using their predictive one-step acquisition probability, adjusted by graph-based target relevance. A simulation-based proof-of-concept study evaluated TEKS on an eight-node symmetric branching graph with two isomorphic target conditions, a six-node asymmetric converging graph, and a six-node linear prerequisite chain. Each condition involved 200 virtual learners, 90 interaction steps, and 30 independent repetitions. TEKS was compared with Prerequisite-Weighted Mastery Gap (PW-MG), a deterministic static schedule, and random selection under matched simulation conditions. In the branching structures, TEKS yielded higher final target mastery, mean target mastery over time, and final scope mastery than all three comparison strategies. In the linear prerequisite chain, TEKS and PW-MG showed similar target-level outcomes, whereas TEKS retained higher final scope mastery. Within the examined conditions, the difference between TEKS and PW-MG was most evident when multiple prerequisite-related candidates competed for selection. These results provide controlled proof-of-concept evidence for target-adjusted one-step acquisition scoring. Further evaluation is needed on larger and more varied prerequisite structures, under broader forms of model misspecification, and with authentic learner interaction data. Full article

Autonomous vehicles increasingly require capabilities that extend beyond perception towards contextual understanding, semantic reasoning, and explainable decision-making. Knowledge graphs (KGs) have emerged as a promising solution by integrating heterogeneous sensor data, traffic regulations, domain knowledge, and contextual information into unified semantic frameworks. This review systematically examines knowledge graph-based intelligent reasoning in autonomous driving through a PRISMA 2020-guided analysis of 47 peer-reviewed studies identified from the literature published from 1 January 2018 to 31 January 2026. The findings reveal that semantic scene understanding and ontology-based representations currently dominate the field, with 66.0% of studies integrating knowledge graphs with deep learning approaches. Neuro-symbolic methods and explainable AI components were identified in 38.3% and 34.0% of publications, respectively, indicating increasing research interest in hybrid and transparent AI architectures. The analysis further demonstrates that 80.9% of studies remain limited to benchmark datasets and simulation environments, whereas only 19.1% provide real-world validation, suggesting relatively low technological maturity and limited industrial readiness. Although KG-enabled approaches substantially improve contextual awareness, hidden hazard anticipation, and explainability compared with conventional perception-centric architectures, major challenges remain regarding scalability, ontology interoperability, semantic error propagation, real-time reasoning, and certification requirements. The review identifies the convergence of knowledge graphs, large language models, and neuro-symbolic AI as a promising direction for next-generation autonomous driving systems. Future research should therefore focus on uncertainty-aware reasoning, adaptive explainability, standardised evaluation methodologies, and certification-oriented real-world deployment strategies. Full article

Machine unlearning (MU) is increasingly studied across artificial intelligence systems in which target information may remain in model weights, adapters, retrieval indexes, caches, federated states, and temporally derived representations. However, existing reviews often examine methods, benchmarks, or deployment settings separately. Therefore, this scoping review collected and synthesized 159 sources to compare unlearning evidence across these connected system components. The evidence map separates 42 core MU sources—26 primary empirical or theoretical studies, 12 benchmark or evaluation frameworks, and 4 secondary evidence syntheses—from 117 contextual sources. First, a three-dimensional taxonomy distinguishes guarantee or reference type, update mechanism, and deployment or memory context. This structure prevents partitioning from being treated as a guarantee and influence estimation from being treated as an outcome. Next, the quantitative map identifies 13 centralized or general sources, 16 LLM- or benchmark-focused sources, 3 graph sources, 3 federated sources, 2 temporal sources, 2 quantized-network sources, and 3 cross-domain reviews. These results indicate a stronger reference-based foundation for centralized MU and a benchmark-rich but transformation-sensitive evidence base for large language models. In contrast, graph, federated, and temporal unlearning remain less mature because they are supported by smaller core evidence sets. Moreover, recent studies show that apparent forgetting may fail after 4-bit quantization, probabilistic decoding, alternative reference selection, recovery testing, benign query changes, or overlap between forget and retain knowledge. Accordingly, we present an audit-oriented deletion lifecycle that connects forgetting evidence, retained-utility testing, threat-model-specific attacks, post-transformation verification, and redeployment decisions. Calibration, citation grounding, provenance, and human review are treated as supplementary decision-readiness checks rather than direct proof of unlearning. Finally, the lifecycle is presented as a structured synthesis and reporting framework that still requires prospective validation in real systems and independent practitioner assessment. Full article

This systematic literature review examines the application of artificial intelligence (AI) and natural language processing (NLP) techniques in intelligent business document processing. The study systematically analyses 46 peer-reviewed articles published between 2014 and 2025 and indexed in the Scopus database. The reviewed literature was grouped into six core NLP-based analytical tasks: semantic search, question answering, summarisation, text data integration and matching, event extraction, and business process management. The findings show that AI- and NLP-based methods have significantly improved the automation, retrieval, interpretation, and structuring of business documents. Semantic search methods enhance information retrieval by moving beyond keyword matching, while question-answering systems and summarisation techniques support automated knowledge discovery and content reduction. Deep learning and transformer-based models have also improved entity matching, event extraction, and predictive business process monitoring. However, the review identifies several persistent limitations, including the continued dominance of extractive approaches, limited adoption of abstractive summarisation, insufficient integration of knowledge graphs, fragmented system development, limited enterprise-scale validation, and a lack of reusable code and shared resources. The findings further indicate that large language models (LLMs), particularly when combined with prompt engineering, retrieval-augmented generation, knowledge graphs, and agent-based architectures, offer promising opportunities to address these gaps. Overall, this review highlights both the progress and remaining challenges in developing scalable, explainable, and domain-adaptable AI-driven systems for intelligent business document processing. Full article

Territorial ecological restoration involves a continuous chain of tasks, including degradation diagnosis, target identification, zoning-based governance, process monitoring, effectiveness assessment and adaptive management. The relevant knowledge is widely distributed across policy plans, monitoring reports, restoration cases and the academic literature, and is characterized by multi-source heterogeneity, cross-scale associations and dynamic change. Existing knowledge organization approaches mainly rely on textual synthesis, indicator systems or general-purpose knowledge graph construction tools, and therefore struggle to simultaneously handle cross-context implicit relations, domain-rule constraints, inconsistent entity expressions and evidence traceability in ecological restoration knowledge. To address these limitations, this paper proposes EcoRestore-KG, a multi-agent knowledge graph construction framework for territorial ecological restoration. The framework unifies heterogeneous inputs through controlled evidence representation and adaptive context segmentation, and organizes ontology-guided triple mining, cross-context relation inference, graph quality control, entity canonicalization, relation endpoint remapping and evidence binding into a progressive workflow for the automatic extraction, auditing and assembly of ecological restoration knowledge. Experimental results show that EcoRestore-KG outperforms general-purpose large language models and existing knowledge graph construction baselines in relation extraction, entity coverage and semantic-quality evaluation. It achieves relation precision, recall and F1 scores of 71.4% ± 0.3%, 69.5% ± 4.4% and 70.3% ± 2.3%, respectively, improving relation F1 by 9.9 percentage points over the strongest baseline. Its entity F1 reaches 79.8% ± 0.7%, and its LLM-S score reaches 8.48 ± 0.11. Single-module and combined ablation experiments further demonstrate that evidence representation, context segmentation, cross-context relation inference, relation quality auditing and entity canonicalization jointly support the performance gains of the framework. This study provides a verifiable methodological pathway for structured organization, quality auditing, evidence tracing and subsequent integration of newly available knowledge. Full article

Industrial IoT maintenance often stops at health prediction, leaving maintenance, rescheduling, safety, and communication to separate decision processes. This study presents MIRA, a safety-constrained graph-based multi-agent reinforcement learning architecture for joint prescriptive maintenance, production rescheduling, and event-triggered communication. Machine condition was estimated from CNC milling data using temporal convolutional models; because predictive uncertainty failed a predefined validation gate, the controller used deterministic health estimates. Evaluation covered five controllers, six simulated scenarios, and 1800 matched episodes. Relative to Graph-MAPPO, MIRA reduced operational cost by 9.38%, weighted tardiness by 28.10%, unexpected failures by 17.39%, message count by 84.98%, and transmitted data by 83.83%, while increasing on-time completion by 23.55%, without a detectable difference in corrected critical-message recall. Across the three independently trained seeds, failures, safety violations, and message count favored MIRA consistently, whereas cost and tardiness favored MIRA in two seeds. Disabling the execution shield increased safety violations from 0 to 3.56 per episode. Post-training variation in the projected-health safe-start threshold from 0.124 to 0.132 produced no safety violations and only small changes in aggregate operational outcomes. Cross-domain health transfer to PHM 2010 failed without adaptation. The results support simulator-level decision coordination, while broader replication, variable-size deployment, and factory validation remain necessary. Full article

Spatiotemporal deep learning models, like Graph Neural Networks (GNNs), Transformers, and selective State Space Models (Mamba), have achieved impressive performance in electroencephalogram (EEG) decoding and affective computing. However, their generalization performance often degrades severely under subject-independent Leave-One-Subject-Out (LOSO) cross-validation protocols. This generalization drop is often attributed to generic domain shifts in standard Brain–Computer Interface (BCI) literature, and is tackled by parameter-heavy adaptations. In contrast, this paper proposes a unified mathematical diagnostic framework to audit and measure the underlying representation and geometric collapse in spatiotemporal brain–computer interfaces. More concretely, we formalize: (1) Topological over-smoothing under volume conduction through Graph Dirichlet Energy bounds indicating GCNs as low-pass filters that smooth localized electrode variations; (2) representation collapse through the Normalized Rank Uniformity Index (NRUI) based on the Shannon Entropy of latent covariance eigenvalues, that distinguishes between dimensional and semantic collapse; and (3) geometric manifold distortions under subject domain shifts on the Symmetric Positive Definite (SPD) Riemannian manifold under the Affine-Invariant Riemannian Metric (AIRM) projection. Auditing these diagnostic metrics on canonical models across DEAP, DREAMER, and SEED, we demonstrate why standard spatiotemporal architectures suffer from performance collapse in cross-subject configurations. We provide BCI engineers with a tangible mathematical blueprint to design robust, collapse-resistant decoders. Full article

Hyperspectral sensors provide high-dimensional spectral–spatial observations for land-cover analysis, but reliable classification remains difficult when only a few target-scene labels are available. Cross-domain few-shot hyperspectral image classification usually depends on sparse support samples, so final decisions can be unstable in low-margin regions where repairable errors and correctly classified long-tail samples are entangled. We propose Boundary-Risk Graph Calibration (BRGC), a risk-controlled calibration framework that improves support-set decision reliability. BRGC combines boundary-aware mixability training with inference-time graph residual calibration. During inference, support labels are clamped, an unlabeled target-query graph provides structural smoothing evidence, and the original classification scores are modified only through low-margin gating and bounded residual updates. On 10 target datasets with 10 random seeds, BRGC consistently improves its base classifier and achieves the highest macro-average OA, AA, and Kappa among matched-protocol transductive baselines. Repair/damage diagnostics, ablation studies, parameter sensitivity analysis, and cross-method adaptation show that BRGC improves scarce-label hyperspectral image interpretation by converting query-graph structure into low-intervention reliability evidence. Full article

Road-network extraction from very high-resolution (VHR) remote-sensing imagery remains a challenging task owing to the structural sparsity, topological complexity, and severe occlusions of road networks. Conventional graph-based approaches preserve topological consistency yet incur considerable computational overhead, whereas prevailing convolutional neural network (CNN) and Transformer architectures struggle to reconcile long-range contextual modeling with computational efficiency. To address these limitations, this study proposes RFM-UNet, a hybrid frequency and state–space network designed for road-network segmentation. Specifically, the encoder integrates Mamba blocks with an Anisotropic Directional Attention (ADA) module to jointly capture local geometric cues and global dependencies at linear computational complexity. In addition, a Multi-Scale Adaptive Fusion Module (MAFM) is introduced to dynamically recalibrate multi-stage features, thereby suppressing cross-scale interference and preserving the connectivity of narrow roads. To enhance robustness against shadow-induced occlusions, a Dual-Spectrum Aggregation Module (DualSpec) decouples the phase and amplitude spectra in the frequency domain and fuses them with spatial features, effectively mitigating spurious responses and background noise characterized by similar textures. Quantitative and qualitative experiments on three public datasets demonstrate that RFM-UNet consistently outperforms current state-of-the-art methods. Full article

Infrared object detection from unmanned aerial vehicles (UAVs) is critically challenged by multi-type composite degradation—including noise, blur, and low contrast—which severely undermines feature discriminability and multi-scale target perception. This paper proposes SGW-DETR (Spectral-Guided Graph-structured Wavelet Detection Transformer), which is a novel framework built upon RT-DETR, incorporating three synergistic modules across the backbone, neck, and encoder. FDSANet (Frequency Domain Spectral Awareness Network) replaces the conventional ResNet backbone, integrating the Multi-Scale Frequency Perception Module (MSFPM), Selective Channel Frequency Decomposition (SCFD), and Dynamic Kernel Spectral Modulation (DKSM) to achieve instance-level adaptive spectral feature extraction without degradation-type supervision. The Graph-Structured Fusion Network (GSFN) combines the Adaptive Semantic Fusion Module (ASFM) with the Graph Structure Perception Module (GSPM), employing Gaussian kernel soft membership and two-stage message passing to explicitly model spatial topological dependencies among object components. The Wavelet-guided Contrast Feature Aggregation module (WCFA) restructures the Attention-based Intra-scale Feature Interaction (AIFI) encoder via a Haar-based Frequency Decomposition Unit (HFDU), decomposing features into foreground-edge and background-thermal components and achieving hierarchical foreground–background decoupling through nested dual-path causal contrastive attention. A UAV infrared degradation dataset comprising 4686 images spanning six degradation types with component-level annotations was constructed for evaluation. SGW-DETR achieves 75.2% mAP₅₀, outperforming RT-DETR by 3.5%, while simultaneously reducing GFLOPs and parameter count by 16.8% and 9.9% at an inference speed of 85.5 FPS. Sustained performance gains on the M3FD and IndraEye benchmarks further demonstrate the framework’s cross-domain generalization capability, offering practical value for UAV-based surveillance, search-and-rescue, and border monitoring under adverse imaging conditions. Full article

Accurate traffic state estimation (TSE) for port motorway networks is critical for mitigating congestion and improving operational efficiency in port cities. Port motorways have their specific traffic patterns, such as surges in container truck demand triggered by vessel arrival events. Considering this characteristic, we propose a novel TSE method named HGCN-VA, enabling a real-time inductive inference based on a graph neural network (GNN). The framework constructs a heterogenous graph comprising both road sensors and ports, while introducing an influence weight learning module based on attention mechanisms to adaptively align vessel events with their traffic consequences. A hierarchical spatial module is further designed, where initial heterogeneous diffuse graph convolution layers explicitly model cross-domain interactions between port and sensor nodes, followed by standard diffusion graph convolution layers to capture traffic propagation patterns. Extensive experiments on a simulation port dataset demonstrate that HGCN-VA outperforms baselines in TSE accuracy. Ablation studies further validate the effectiveness of incorporating vessel information and the proposed heterogeneous spatial modeling strategy. Full article