Search Results (3,442)

Detecting small anomalous targets in hyperspectral imagery is challenging when neither target spectra nor spatial morphologies are known, because most unsupervised detectors still encode implicit assumptions about background statistics or target structures and therefore generalize poorly to genuinely unknown threats. This paper introduces the Hard-Negative-Guided Topology Transfer Network (HNGT-Net), a lightweight teacher–student framework that addresses this problem from two perspectives. On the representation side, a similarity graph coupling spatial adjacency with feature-space nearest neighbors characterizes the spectral–spatial topology of normal backgrounds, and a three-level consistency objective over node embeddings, structural relations, and graph-Laplacian responses transfers this topology from a frozen teacher to a compact student. On the discrimination side, hard negatives are synthesized directly on pure normal pixels through sparsity-gated projected gradient perturbation, while a response-margin constraint forces the student to score such negatives above normal samples and mitigates distribution overfitting. Experiments on four public benchmark scenes and one Salinas-derived synthetic dataset show that HNGT-Net achieves an average AUC of 0.9953 with the smallest variance among all competitors. The student branch provides 10.5-fold parameter compression relative to the teacher, while the complete teacher–student stack remains compact, supporting resource-constrained remote sensing deployment. Full article

Weather radar calibration is essential for ensuring data consistency and quantitative precipitation estimation in X-band radar networks. Existing absolute calibration methods (e.g., metal sphere, horn antenna) suffer from high cost, poor timeliness, and difficulty in automation due to meteorological conditions and airspace restrictions, while spatiotemporal matching methods based on volume scan data suffer from interpolation and matching inaccuracies. To address these issues, this study proposes a collaborative calibration method for X-band radar networks based on opposing Range–Height Indicator (RHI) scans. The method uses a rigorously calibrated reference radar as a benchmark and performs opposing RHI scans with the radar under calibration to obtain synchronized observations within the spatial overlap region. Precise spatial matching is achieved using the nearest-neighbor algorithm based on beam-broadening cross-coverage thresholds, and bias is extracted using both the midline 9-point averaging method (midline method) and spatially constrained regional Statistics method (regional method). Based on a total of 58 sets of opposing RHI scanning cases conducted under stratiform precipitation, scattered precipitation, and weak cloud conditions, the results show that under conditions where echo continuity is maintained near the midline of stratiform and scattered precipitation, both the midline method and the regional method can obtain stable matching data. The midline method achieves a median correlation coefficient (0.821–0.942) higher than that of the regional method (0.860–0.872), and its bias standard deviation remains relatively stable (midline method: 1.39–2.20 dB; regional method: 2.61–3.16 dB). Continuous RHI calibration tests confirm that within a 30-min window, the fluctuation of the data matching correlation coefficient is less than 0.05, and the fluctuation of the bias mean is controlled within ±0.3 dB. Under weak cloud conditions, although the midline method can still achieve a high correlation coefficient, the correctness of its results still requires auxiliary validation through other calibration means. This study provides a relatively efficient and effective technical approach for the automated collaborative calibration of dense X-band radar networks. Full article

Class imbalance, where majority-class samples vastly outnumber minority-class samples, remains a persistent challenge in network intrusion detection systems (NIDS), often causing classifiers to overlook rare but critical attack types while yielding misleadingly optimistic performance metrics. Synthetic data generation is a common mitigation strategy. However, existing methods often fail to capture the non-linear network traffic and neglect inter-class relationships with the majority class, resulting in inconsistent performance gains. This study investigates three contrastive learning loss functions, Contrastive Loss, Soft Nearest Neighbor Loss, and Supervised Contrastive Loss, integrated into a Conditional Variational Autoencoder (CVAE) regularised via the standard Kullback-Leibler divergence objective. Experiments were conducted on four widely used NIDS benchmark datasets (NSL-KDD, UNSW-NB15, CIC-IDS2017, and CSE-CIC-IDS2018), with synthetic data evaluated using four machine learning classifiers against the original imbalanced data, a non-contrastive CVAE, and conventional oversampling approaches. The results show that the effectiveness of integrating contrastive learning into the CVAE framework is dependent on the specific dataset, minority class, contrastive loss function, and distance metric, with the proposed approach outperforming traditional oversampling techniques in several settings without degrading majority-class performance or overall accuracy. These findings provide practical guidance for selecting contrastive learning objectives in class-imbalanced NIDS scenarios. Full article

Background: Long COVID has emerged as a global health challenge characterized by persistent neuropsychiatric symptoms, including depression, fatigue, and cognitive impairment. Growing evidence indicates that sustained neuroinflammation, endothelial dysfunction, and fronto-limbic network disruption may underlie these symptoms, representing pathophysiological features distinct from major depressive disorder (MDD). Although repetitive transcranial magnetic stimulation (rTMS) is an established treatment for MDD, its therapeutic efficacy in Long COVID-associated depression remains uncertain. Methods: Long COVID was defined as laboratory-confirmed SARS-CoV-2 infection followed by persistent neuropsychiatric symptoms lasting ≥3 months, including cognitive impairment (“brain fog”), fatigue, and new-onset depressive symptoms. We conducted a retrospective registry-based cohort study using real-world clinical data from two TMS clinics in Tokyo (May 2022–April 2026). Forty-four medication-free patients with Long COVID-associated MDD were compared with eighty-eight propensity score–matched patients with primary MDD (1:2 nearest-neighbor matching based on age, sex, and baseline Montgomery–Åsberg Depression Rating Scale (MADRS)). All participants received left dorsolateral prefrontal cortex intermittent theta burst stimulation (iTBS). Treatment outcomes were evaluated using MADRS and 17-item Hamilton Depression Rating Scale (HAM-D₁₇) improvement rates, HAM-D₁₇ response, and remission. Statistically adjusted analyses (inverse probability of treatment weighting (IPTW) and doubly robust estimation) were used to reduce measured confounding; however, residual unmeasured confounding cannot be excluded due to the retrospective observational design. Results: Long COVID patients showed significantly attenuated antidepressant response. Improvement rates were markedly lower for MADRS (38.4% vs. 61.5%) and HAM-D₁₇ (36.7% vs. 58.4%). IPTW and doubly robust models demonstrated large adjusted percentage-point differences (MADRS: −23.0 percentage points; HAM-D₁₇: −21.8 percentage points; both p < 0.001). While HAM-D₁₇ response did not differ significantly, remission rates were substantially reduced (27.3% vs. 61.4%). Conclusions: These findings indicate an adjusted association suggesting reduced rTMS responsiveness in Long COVID-associated depression. Given the retrospective observational design and the possibility of residual unmeasured confounding, the results should be interpreted as associations rather than evidence of a causal effect. Full article

Accurately identifying influential nodes is essential for analyzing network structures and optimizing information propagation. Existing methods predominantly rely on single indicators such as degree, H-index, or k-shell, inherently limiting their ability to capture a node’s true influence. Recent hybrid centrality approaches attempt to address this by combining multiple local and global attributes; however, they typically integrate features through simple weighting or superposition, failing to characterize the intrinsic synergy among structural properties. Furthermore, they often quantify neighbor contributions too coarsely, overlook the regulatory role of edge strength, and some suffer from high computational complexity, limiting scalability. To overcome these deficiencies, we propose WKDH, a novel influential node identification method based on multi-attribute neighbor contributions. WKDH fuses local structural attributes (degree and H-index) with global structural attributes (k-shell) via a multiplicative weighted synergy model, simultaneously capturing local connection “quantity,” local connection “quality,” and global core-layer position. By transforming neighbors’ comprehensive characteristics into regulated contribution degrees, WKDH mitigates excessive self-attribute interference and accurately reflects the actual propagation potential of edges. Notably, the method achieves linear computational complexity of $O\left(m\right)$. Experimental results on nine real-world and six artificial networks demonstrate that WKDH outperforms nine established indicators in terms of node influence ranking, identification of high-influence nodes, and measuring propagation capability. Moreover, WKDH exhibits strong universality across diverse network structures, as it operates without parameter tuning. Full article

Solanum rostratum is an invasive plant species that poses a serious threat to native ecosystems. In this study, we investigated S. rostratum populations in saline-alkali regions of western Jilin Province to characterize root-associated arbuscular mycorrhizal (AM) fungal diversity and evaluate the effects of AM fungi on plant growth and competitive ability. The results showed that S. rostratum roots were widely colonized by AM fungi, indicating abundant AM fungal resources in its rhizosphere. High-throughput sequencing identified 45 AM fungal species belonging to 13 genera, with Glomus as the dominant genus. A compartmented mesh pot experiment further demonstrated that AM fungal inoculation significantly enhanced the growth and photosynthetic performance of S. rostratum, particularly under saline-alkali soil conditions. When AM fungi were inoculated only in the compartment containing the native plant Setaria viridis, AM fungal colonization was also detected in S. rostratum roots, indicating hyphal connections between neighboring plants through the mesh barrier. Stable isotope analysis further revealed that AM fungi facilitated nitrogen acquisition by S. rostratum and increased ¹⁵N transfer from neighboring S. viridis under common mycorrhizal networks. These findings suggest that abundant AM fungal resources in the rhizosphere contribute to the growth advantage of S. rostratum. By enhancing nutrient acquisition and potentially facilitating nitrogen transfer through mycorrhizal networks, AM fungi may strengthen the competitive ability of this invasive plant under saline-alkali conditions. Full article

Against the background of global production-network restructuring, low-carbon transition, and rising external uncertainty, this study examines the effect of green infrastructure investment on urban industrial chain resilience. Using panel data for 285 Chinese prefecture-level cities from 2012 to 2024, industrial chain resilience is measured from the dimensions of industrial diversification and urban innovation capacity. Double machine learning is employed for baseline estimation, supplemented by mediation analysis, threshold regression, spatial econometric analysis, and a series of robustness tests. The results show that green infrastructure investment significantly enhances industrial chain resilience, and the finding remains robust to alternative model specifications, cross-fitting settings, generalized propensity score weighting, continuous-treatment entropy balancing, winsorization, and the exclusion of pandemic-period observations. Resource allocation efficiency plays a partial mediating role in this relationship. The threshold analysis identifies a significant nonlinear effect associated with energy consumption intensity, with the positive effect of green infrastructure investment being stronger below the estimated threshold and weakening above it. Spatial analysis further shows significant spatial dependence in both green infrastructure investment and industrial chain resilience, together with positive spillover effects on neighboring cities. These findings highlight the importance of improving green infrastructure investment efficiency, strengthening factor allocation, and promoting regional coordination in enhancing urban industrial chain resilience. Full article

This paper studies a risk-averse stochastic unit commitment framework for an energy aggregator, operating a portfolio of conventional generators, renewable units, and battery energy storage in a network-constrained environment. Renewable generation and demand uncertainty are represented through a scenario-based extensive-form mixed-integer linear program. To avoid exposure to rare but high cost events, the model incorporates conditional value-at-risk as part of the objective function. The approach captures key market interactions, including day-ahead commitments, imbalance penalties, and power exchange with a neighboring network, while respecting generator constraints, storage dynamics, line flow limits, and bus voltage security. A comprehensive parametric study is conducted to quantify the influence of two risk parameters: the conditional value-at-risk confidence level $\alpha$ and the risk-aversion weight $\lambda$. Using a 300-scenario test set on a modified IEEE 9-bus system, the results show that risk-neutral scheduling exposes the aggregator to larger operational costs in extreme scenarios. Minor levels of risk aversion (0.1–0.5) reduce CVaR and tighten the distribution of costs. Increasing $\lambda$ further yields diminishing returns, while higher $\alpha$ values focus risk mitigation on the most severe outcomes. The results demonstrate how CVaR-based stochastic scheduling can support aggregator decision-making by quantifying downside risk under renewable uncertainty. Full article

Multi-agent reinforcement learning has emerged as a promising paradigm for cooperative unmanned aerial vehicle (UAV) swarm coordination. However, existing communication-aware MARL methods primarily focus on communication topology, message routing, and message aggregation, while the information content of the exchanged messages is often only implicitly controlled. In realistic UAV networks, inter-agent communication is constrained by limited bandwidth, communication range, energy consumption, and packet loss. It is therefore desirable for each UAV to transmit compact and task-relevant information rather than dense and redundant latent features. In this paper, we propose IB-CEMARL, an information-bottleneck-guided, communication-efficient multi-agent reinforcement learning framework for UAV swarms. We formulate inter-UAV communication as a minimal sufficient message-learning problem in which each UAV encodes its local observation into a stochastic bottleneck message before exchanging information with its neighbors. Cauchy–Schwarz divergence-based quadratic mutual information is adopted as a unified dependence measure to jointly regularize message compression, preserve decision-relevant information, and reduce statistical redundancy among neighboring UAV messages. Extensive experiments demonstrate that IB-CEMARL achieves superior cooperative performance, reduced message redundancy, and stronger robustness compared with representative communication-aware MARL baselines. In particular, IB-CEMARL improves the average return by $4.9\%$ and reduces inter-message dependence by $29.0\%$ compared with the KL-IB-MARL baseline while maintaining efficient communication under constrained bandwidth settings. Full article

Rare-event detection is a critical task in intelligent risk monitoring systems, where missed minority events may lead to financial loss, security threats, or operational failures. Existing imbalance-handling methods frequently apply one class-level correction and therefore ignore the heterogeneous geometric roles of minority observations. This study presents density–boundary graph reweighting (DBGR), a unified sample-level weighting framework that combines continuous minority sparsity and majority-boundary exposure and regularizes the resulting scores through a minority-only K-nearest-neighbor graph. Its methodological contribution lies in this joint pre-training formulation and in producing classifier-compatible weights, rather than in claiming novelty for density estimation, graph propagation, or weighting individually. Two variants are considered: DBGR-Safe for sparse and relatively safe minority prototypes and DBGR-Danger for sparse boundary observations. Experiments on financial fraud detection, industrial fault diagnosis, and network intrusion detection show recall-oriented gains with XGBoost and reduced variability for the tested LightGBM setting, while the benefit is limited for Random Forest. On Creditcard, DBGR-Danger improves Recall from 0.8223 to 0.8949 and F2 from 0.8410 to 0.8931. On NSL-KDD U2R, focal loss remains better in Recall, F2, and PR-AUC, whereas DBGR-Safe achieves the best F1. DBGR is therefore positioned as a complementary, classifier-compatible strategy rather than a universally superior imbalance-handling method. Full article

Student retention analytics often treats withdrawal as a single outcome, although students may attribute dropout intention to personal, socioeconomic, and academic pressures simultaneously. We benchmarked nine supervised classifiers for identifying a concurrent three-domain attribution profile in a cross-sectional survey of 333 undergraduates at a Colombian public university campus. The response came from a semi-structured weight-allocation item; an audit found that literal label matching altered 21 classifications because of spelling variants and decimal notation. Nine classifiers—logistic regression, decision tree, random forest, neural network, Gaussian Naïve Bayes, k-nearest neighbors, AdaBoost, gradient boosting, and XGBoost—were fitted using six pre-specified predictors. Models were compared by repeated nested stratified cross-validation (five outer folds, three repeats), inner tuning, fold-contained preprocessing, and training-only threshold selection. The concurrent profile occurred in 256 students (76.9%). Logistic regression achieved the highest mean held-out ROC AUC (0.674, 95% CI 0.644–0.704), closely followed by random forest (0.671, 0.641–0.702); their paired difference was nonsignificant after Holm adjustment. Logistic regression had the highest F1 score (0.788), whereas random forest had the highest balanced accuracy (0.617). AdaBoost did not retain its apparent single-holdout advantage. Housing and financial aid had the largest held-out permutation importance. The predictors provided moderate discrimination of a perceptual profile, not a validated prediction of future dropout. Outcome auditing and leakage-free validation materially changed the model ranking. Full article

Remote-sensing-derived spectral indices and in situ measurements provide complementary information for aquatic monitoring, but their practical integration is complicated by asynchronous observations, heterogeneous tables, missing acquisitions, and non-reproducible preprocessing. This study presents AI-Aquatica-RS, a modular Python framework for spectral-index calculation, station-aware nearest-neighbor temporal alignment, feature-set construction, regression benchmarking, command-line execution, and structured result export. The software was evaluated using a fully synthetic controlled benchmark; no real satellite scenes or field-monitoring measurements were used. The benchmark comprised 1080 daily in situ-like observations from six stations and 181 unique remote-sensing-like acquisitions generated as water-like surface-reflectance proxies. A ±3-day alignment tolerance produced a shared complete-case cohort of 954 records. To ensure a fair comparison, the in situ-only, spectral-index-only, and fused configurations were evaluated on exactly the same 667 training and 287 validation records. The fused configuration achieved the best performance using ridge regression (RMSE = 3.481 NTU, MAE = 2.768 NTU, R² = 0.729), compared with RMSE values of 5.086 NTU for the in situ-only configuration, and 5.538 NTU for the spectral-index-only configuration. The benchmark demonstrates reproducible execution and recovery of complementary information under controlled conditions; it does not constitute environmental validation. AI-Aquatica-RS provides an extensible software layer for future studies using real satellite products, monitoring networks, sensor-specific preprocessing, and spatially blocked validation. Full article

To address the challenges of integrating multi-source heterogeneous data, fragmented fault knowledge, and the limited capability of traditional rule engines in recognizing edge cases for ultra-high voltage (UHV) bushing fault diagnosis, this paper proposes a fault identification and decision-making method based on knowledge graph (KG) rule reasoning and inductive graph convolutional network (Inductive GCN). First, a triple-matching strategy is employed to perform entity extraction and relation mining from fault cases, constructing a fault knowledge graph that transforms unstructured fault case texts into a structured knowledge graph. Second, a rule engine based on a multi-source feature rule set is designed, utilizing the entropy weight method and the RETE algorithm to achieve interpretable symbolic reasoning. On this basis, a double-layer inductive graph convolutional network is introduced to learn implicit fault patterns by aggregating topological information from neighboring nodes, and a confidence-driven dynamic weighted fusion strategy is adopted to achieve complementary advantages between the two models. Finally, a large language model is introduced to generate operation and maintenance decision recommendations. Experimental results demonstrate that the proposed method achieves an identification accuracy of 98.1% on a test set of 159 samples, which is 10.7 percentage points higher than that of a single rule engine and 6.9 percentage points higher than that of a single inductive graph convolution network. The standard deviation of accuracy across different test batches is only 0.0029. These results demonstrate the effectiveness and stability of the proposed method, providing a practical technical solution for UHV bushing fault identification. Full article

Mobile LiDAR systems (MLS) enable rapid acquisition of large-scale 3D point cloud data. Semantic segmentation of the acquired point clouds is an important task in outdoor scene understanding and environmental perception for autonomous driving. However, existing methods tend to suffer from boundary confusion when segmenting MLS point clouds with long-tail categories. To address this problem, we propose the Dilated Context Attention Network (DCA-Net), which consists of a dilated local geometric encoding module, a channel attention pooling module, and a category-boundary sampling strategy. The dilated local geometric encoding module expands point-to-point connections within a fixed neighborhood to strengthen contextual modeling among neighboring points. The channel attention pooling module uses a channel attention mechanism to enhance informative channel responses in neighborhood features, thereby improving local feature representation. The category-boundary sampling strategy increases the sampling probabilities of minority-category points and boundary points, reducing feature information loss during down-sampling. Experimental results on the S3DIS, Toronto3D, and MLS road scene datasets show that DCA-Net achieves mIoU scores of 69.4%, 84.1%, and 96.6%, respectively. These results demonstrate that the proposed method alleviates boundary confusion in point cloud segmentation with long-tail categories, without causing a noticeable degradation in the segmentation performance of majority categories. Full article

Soil organic carbon (SOC) is a key indicator of cropland quality, soil fertility, and the carbon sequestration potential of agroecosystems. Accurate characterization of its spatial distribution is essential for black soil conservation and regional soil carbon management. However, regional-scale SOC prediction is often constrained by limited field observations, which can reduce model generalizability and predictive reliability. In this study, we developed a limited-sample SOC prediction framework for the Liaohe Plain using 310 surface (0–20 cm) soil samples collected in 2025 and multi-source environmental covariates, including climate, vegetation, soil spectral, and terrain variables. The framework used the Tabular Prior-Data Fitted Network (TabPFN), whose performance was compared with that of Random Forest, Support Vector Machine, CatBoost, K-Nearest Neighbors, and XGBoost. Model performance was evaluated using 100 repetitions of random 80:20 holdout validation and repeated five-fold spatial cross-validation based on spatially constrained clustering, while sampling-induced relative uncertainty was quantified using 100 repeated random sampling and model-fitting runs. Under random holdout validation, TabPFN showed competitive predictive performance, with mean R² and RMSE values of 0.608 ± 0.038 and 4.026 ± 0.184 g kg⁻¹, respectively. Repeated spatial cross-validation yielded more conservative performance estimates, with mean R² and RMSE values of 0.535 ± 0.072 and 4.33 ± 0.31 g kg⁻¹, respectively, indicating that random splitting may overestimate model performance when sampling sites are spatially clustered. Spatial prediction showed that cropland SOC ranged from 4.63 to 27.04 g kg⁻¹, with generally lower values in the west and higher values in the northeast. Areas with high sampling-induced relative uncertainty were mainly concentrated in the northern, northeastern, and marginal regions. These findings provide a methodological basis for SOC mapping, supplementary sampling optimization, and regional soil carbon management under limited-sample conditions, although the temporal robustness of the results requires confirmation using independent data from additional years. Full article