<?xml version="1.0" encoding="utf-8"?>
    <rss version="2.0">
      <channel xmlns:content="http://purl.org/rss/1.0/modules/content/">
        <title>Frontiers in Big Data | New and Recent Articles</title>
        <link>https://www.frontiersin.org/journals/big-data</link>
        <description>RSS Feed for Frontiers in Big Data | New and Recent Articles</description>
        <language>en-us</language>
        <generator>Frontiers Feed Generator,version:1</generator>
        <pubDate>2026-07-13T12:22:56.779+00:00</pubDate>
        <ttl>60</ttl>
        <item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fdata.2026.1878260</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fdata.2026.1878260</link>
        <title><![CDATA[A dataset-centric review of IoT and IIoT intrusion detection: realism, evaluation biases, and future research directions]]></title>
        <pubdate>2026-07-10T00:00:00Z</pubdate>
        <category>Review</category>
        <author>Dwarsala Sreedhar Reddy</author><author>Kakelli Anil Kumar</author>
        <description><![CDATA[The rapid growth of IoT and IIoT expands the cyber-attack surface of interconnected and safety-critical systems, and, as such, IDSs have become a fundamental security mechanism. Although very impressive results have been reported for machine learning and deep learning-based IDS in benchmark datasets, these gains often do not generalize to real-world deployments owing to dataset design limitations, realism deficits, and evaluation biases, rather than inherent flaws in detection algorithms, which can lead to significant vulnerabilities in actual operational environments. This study presents a dataset-centric review of widely used intrusion detection datasets from the IIoT, IoT, and traditional network domains. A unified taxonomy differentiates datasets based on the domain context, traffic representation, protocol semantics, and attack modeling assumptions. Based on a common analytical framework, each dataset was reviewed regarding its realism, coverage of the threats, class imbalance, temporal continuity, and modern ML/DL-based evaluation of the IDS. The cross-dataset analysis conducted in this study shows that, in addition to the fact that model architecture and feature engineering play a major role, several studies indicate that the simplicity of the datasets, the class imbalance, and the repetitive attack patterns as well as the evaluation methods can affect accuracy of the IDS. This work further underlines the remaining gaps, such as zero-day and adaptive attacks, limited encrypted traffic, weak temporal evolution, poor support for federated learning, and sparse annotations for explainable IDSs. Finally, this study presents future directions for dataset design aligned with the requirements of next-generation IDSs by highlighting digital twin-based IIoT environments, edge-cloud collaborative data generation, sequential traffic modeling, and explainability-oriented annotations that can ensure robust, trustworthy, and deployment-ready IDS solutions.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fdata.2026.1728498</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fdata.2026.1728498</link>
        <title><![CDATA[A prognostic tool for pulmonary collapse: nomogram-based prediction of 28-day mortality]]></title>
        <pubdate>2026-07-09T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Xinming He</author><author>Wenchong Yu</author><author>Yuling Li</author><author>Ao Ma</author><author>Zhichao Meng</author><author>Jiehao Zhu</author><author>Minghui Tan</author><author>Xiaodong Zhao</author><author>Mu Chen</author>
        <description><![CDATA[BackgroundPulmonary collapse is a common and serious respiratory condition, but there is no dedicated bedside tool to estimate prognosis. This study aimed to develop a nomogram to predict 28-day mortality in patients with pulmonary collapse.MethodsWe extracted data for patients with pulmonary collapse from MIMIC-III, identified predictors using regression analyses, and used MIMIC-IV for temporal validation. We then built a nomogram based on the selected predictors. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), AUC comparisons using the DeLong test, reclassification (NRI and IDI), calibration (calibration curves, calibration slope, and Brier score), and decision curve analysis (DCA).ResultsA total of 4,088 patients with pulmonary collapse were included in the study. Logistic regression analysis identified twelve independent predictive factors associated with 28-day mortality: age (OR = 1.01, P =0.040), married status (OR =0 .63, P =0 .040), Glasgow Coma Scale score (OR = 0.94, P = 0.02), creatinine (OR = 0.82, P = 0.03), chloride ions (OR = 0.85, P = 0.03), sodium ions (OR = 1.19, P = 0.02), blood urea nitrogen (OR = 1.02, P < 0.001), white blood cell count (OR = 1.04, P < 0.001), heart rate (OR = 1.02, P = 0.02), respiratory rate (OR = 1.05, P = 0.03), temperature (OR = 0.54, P < 0.001), and metastatic cancer (OR = 6.66, P < 0.001). The nomogram showed moderate discrimination and consistently higher AUC than Age+Gender, SOFA, and SAPSII across cohorts.ConclusionThis study identified factors associated with 28-day mortality in patients with pulmonary collapse and developed a nomogram for early risk stratification.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fdata.2026.1837706</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fdata.2026.1837706</link>
        <title><![CDATA[Noise-robust temporal–spectral fusion transformers for EEG-based cognitive state classification in aviation environments]]></title>
        <pubdate>2026-07-09T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Quynh Anh Nguyen</author><author>Nam Anh Dao</author><author>Long Nguyen</author>
        <description><![CDATA[Attention-related Pilot Performance Decrements (APPD) contribute substantially to aviation incidents, yet existing electroencephalography (EEG)-based monitoring methods often lack generalization, robustness to noise, and effective temporal–spectral integration. We propose a temporal–spectral fusion transformer (TF-T) combining multi-scale preprocessing, dual-stream temporal and spectral feature extraction, and transformer-based fusion with enhanced temporal–spectral integration and multi-resolution feature processing for multiclass cognitive state recognition. Three variants (TF-T1–TF-T3) are evaluated on controlled and ecologically realistic EEG datasets under clean and noise-augmented (Gaussian, Uniform, COMBO) conditions, using chronological partitioning to avoid temporal leakage. TF-T2 achieves the highest clean-data accuracy (99.2%), while TF-T3 offers superior robustness, improving Macro-F1 by ~4.5–4.7 points across all noise types and outperforming state-of-the-art baselines by up to +8 Macro-F1 under COMBO noise, supporting its deployment in perturbation-prone aviation environments.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fdata.2026.1842233</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fdata.2026.1842233</link>
        <title><![CDATA[Evolutionary multi-agent reinforcement learning for crisis-aware demographic policy optimization]]></title>
        <pubdate>2026-07-08T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Anton V. Dozhdikov</author><author>Arseniy M. Sitkovskiy</author>
        <description><![CDATA[Demographic systems face unprecedented challenges from simultaneous crises. Conventional statistical demography techniques and agent–based models often struggle to capture nonlinear inter–regional interactions during periods of severe socio–economic disruption. To address this, we propose MADDPG–EVO–DGM, a hybrid algorithm that integrates multi–agent deep reinforcement learning with evolutionary optimisation and meta–learning principles to model regional demographic processes under multiple crisis scenarios. Each region is treated as an autonomous agent learning to steer demographic policy levers, while periodic evolutionary “boosters” overcome local optima via population–based perturbations of actor network parameters. Additionally, a Darwin–Gödel Machine–inspired meta–learning mechanism adapts the booster triggers, enabling self–improvement in the learning process. We evaluate MADDPG–EVO–DGM on a simulation environment calibrated with real demographic data for eight federal regions of the Russian Federation over the period 2000–2024 and subject to ten concurrent crisis scenarios (e.g., pandemic, geopolitical conflict, economic collapse). Experiments demonstrate significantly faster convergence and improved performance over a baseline MADDPG: the hybrid approach achieves a higher final average reward (252.57 vs. 243.07) and 3.4 × lower convergence variance (σ = 0.24 vs. 0.80), indicating more reliable training. It also exhibits qualitative performance jumps of +68% during evolutionary phases and maintains 35%–45% greater resilience under crisis shocks compared to the baseline. To our knowledge, this is the first application of multi–agent reinforcement learning to large–scale demographic modeling under crises, opening new possibilities for evidence–based, crisis–resilient population policy design. Code, data, and logs are provided to ensure reproducibility.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fdata.2026.1883452</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fdata.2026.1883452</link>
        <title><![CDATA[Cross-model evaluation of phishing detectors against LLM-generated emails]]></title>
        <pubdate>2026-07-07T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Rommel Gutierrez</author><author>William Villegas-Ch</author><author>Jaime Govea</author>
        <description><![CDATA[Phishing remains a prevalent cyberattack vector, and the widespread adoption of large language models (LLMs) has enabled adversaries to generate grammatically correct and contextually coherent phishing emails at scale, against which conventional detection systems are less effective. Although stylometric methods achieve over 95% accuracy within a single generator, their performance has not been systematically evaluated when the source model changes between training and deployment. This represents a significant gap, as adversaries can switch generators rapidly. A balanced corpus of 9,986 phishing emails was assembled, comprising 4,986 emails generated by three modern LLMs (GPT-4.1, DeepSeek 3.2, and Llama 3.3 70B) across five thematic categories, and 5,000 human phishing emails sampled in a stratified manner from five public sources. Seventeen stylometric features were extracted, and Logistic Regression and XGBoost classifiers were evaluated under intra-model, cross-model, threshold-recalibrated, cross-dataset, and aggregated-pool settings. Intra-model F1 scores reached 0.96 under stratified cross-validation and 0.999 on held-out splits used for the cross-model matrix. However, cross-model F1 dropped by 28.0 percentage points under the default decision threshold of 0.5. Notably, the area under the receiver operating characteristic curve remained above 0.96 in every off-diagonal cell, indicating that discriminative information is preserved even though the decision threshold is generator-specific. Recalibrating the threshold on a small target subset reduced the gap to 4.0 percentage points (an 86% reduction), and an aggregated-pool detector achieved F1 = 0.997 on each generator. This work reframes cross-model phishing detection from a problem of model incompatibility to one of practical calibration, and provides two deployable solutions, threshold recalibration on a small target slice and aggregated-pool training, along with a publicly released multi-LLM corpus.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fdata.2026.1871346</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fdata.2026.1871346</link>
        <title><![CDATA[Adaptive class-aware feature selection for high-dimensional and imbalanced multi-class network intrusion detection]]></title>
        <pubdate>2026-06-23T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Joseph P. Mchina</author><author>Neema Mduma</author><author>Ramadhani S. Sinde</author>
        <description><![CDATA[High-dimensional feature spaces and severe class imbalance remain fundamental challenges for Machine Learning-based Network Intrusion Detection Systems (ML-NIDS), where minority attack categories are frequently overlooked during feature selection. Existing feature selection approaches commonly rely on global feature relevance measures and manually specified feature counts, which favor majority traffic classes and reduce sensitivity toward rare but critical attack categories. To address these limitations, this study proposes the Adaptive Class-Aware Feature Selection (ACAFS) framework for multi-class intrusion detection. Unlike conventional approaches, ACAFS introduces a data-driven adaptive feature count mechanism based on permutation null hypothesis testing, a Class-Aware Composite Mutual Information scoring strategy that explicitly preserves minority-class discriminative information, and a coordinated two-stage feature selection framework that combines statistical filtering with XGBoost-based model refinement. The framework was evaluated independently on the CSE-CIC-IDS2018 benchmark dataset and a Simulated University Network Environment (SUNE) dataset representing Tanzanian higher learning institution networks. Experimental results demonstrate that ACAFS substantially reduces feature dimensionality while improving balanced intrusion detection performance. On CSE-CIC-IDS2018, ACAFS reduced the feature space from 74 to 22 features, representing a 70.3% dimensionality reduction, while the Two-Stage CNN achieved 99.39% accuracy, 99.40% F1-score, and a false positive rate of 0.09%. The framework further achieved 98.59% recall for Web_Attacks despite severe class imbalance, demonstrating effective preservation of minority-class discriminative features. On the SUNE dataset, ACAFS independently selected 18 features and maintained stable detection performance without dataset-specific manual tuning, confirming its adaptability across heterogeneous network environments. These results confirm that adaptive and class-aware feature selection can simultaneously reduce feature redundancy, improve minority attack detection, and maintain robust intrusion detection performance across diverse network traffic environments.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fdata.2026.1807559</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fdata.2026.1807559</link>
        <title><![CDATA[Influence of localized roadway surface obstacles on vehicular emissions under real-world urban driving conditions]]></title>
        <pubdate>2026-06-23T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Victor Cardoso Oliveira</author><author>Thiago Iachiley Araújo de Souza</author><author>Nicole Souza Batista</author><author>Bruno Vieira Bertoncini</author><author>Verônica Teixeira Franco Castelo Branco</author>
        <description><![CDATA[IntroductionVehicular emissions are a major source of air pollution in tropical urban environments. While the impacts of technology, traffic flow, and driving behavior on pollutant formation are well established, the influence of pavement surface remains insufficiently understood. Pavement defects such as potholes, cracks, and depressions disturb vehicle operation and may increase real-world emissions. This study evaluates the influence of pavement obstacles on emissions of CO2, CO, and NOx across five urban road segments in Fortaleza, Brazil.MethodsA portable emissions measurement system (PEMS) collected second-by-second exhaust data during real driving. Roadway surface obstacles were captured through windshield-mounted images acquired at 1 Hz. A broader image pool comprising 44,175 roadway images was assembled from multiple urban roads for model development. From this pool, a final annotated dataset comprising 1,812 images with 3,158 labeled obstacles was used for training, validation, and testing the YOLOv8n detector, which was then applied to the five monitored road sections used in the synchronized emission analysis. Emission data and obstacle locations were synchronized, enabling comparison of pollutant rates along obstacle-present and obstacle-free segments, with emphasis on features likely to influence short-term driving behavior. Detection performance was evaluated using precision, recall, and mAP metrics.ResultsRoad segments with higher obstacle occurrence presented elevated emission rates. In the full dataset, maximum values reached 478.2 g/km for CO2, 491.96 mg/km for CO, and 100.266 mg/km for NOx. A filtered analysis excluding curves, intersection buffers, and visible traffic or pedestrian interference showed that obstacle-present observations still exhibited higher emissions than obstacle-free observations, with average increases of 26% for CO2, 31% for NOx, and 42% for CO. Spatial mapping showed that emission hotspots tended to occur in areas with frequent roadway surface obstacles and operational disturbances.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fdata.2026.1687969</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fdata.2026.1687969</link>
        <title><![CDATA[Deep learning model to predict COPD hospital admissions based on meteorological data: a medical meteorological forecast]]></title>
        <pubdate>2026-06-18T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Lei Zhang</author><author>Mingjie Zhang</author><author>Jinghong Zhang</author><author>Yajie Zhang</author><author>Tian Xie</author><author>Yipeng Ding</author><author>Shuyuan Chu</author><author>Haihong Wu</author>
        <description><![CDATA[BackgroundChronic obstructive pulmonary disease (COPD) has placed a substantial health burden on the world. Meteorological conditions are associated with hospital admissions for COPD. In this study, we aim to develop a model of medical meteorological forecasting for COPD hospital admissions.MethodsA predictive model was developed using a Long Short-Term Memory (LSTM) algorithm applied to time series data on COPD hospital admissions and meteorological conditions. Data were collected daily from 25 September 2016 to 26 December 2020. Performance of the model was assessed using the mean squared error (MSE), root mean squared error (RMSE), mean absolute error (MAE), and R2. The association between the risk of COPD hospital admissions and meteorological conditions was assessed using a conditional logistic regression analysis and a conditional Poisson regression analysis in a time-stratified case-crossover design.ResultsA total of 17,555 hospital admissions for COPD from 1 January 2017 to 31 December 2019 were included in the final LSTM model. Regarding the performance of the LSTM model, the MSE was 0.028, the RMSE was 0.167, the MAE was 0.134, and R2 was 0.416. Regression analysis revealed that the maximum temperature was positively associated with COPD hospital admissions.ConclusionThe LSTM model offers potential for medical meteorological forecasting to predict COPD hospital admissions among the general population according to the local climate. Higher maximum temperature may be a risk factor for COPD hospital admissions.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fdata.2026.1857064</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fdata.2026.1857064</link>
        <title><![CDATA[Where diverse populations gather: transit accessibility and the spatial structure of social mixing]]></title>
        <pubdate>2026-06-18T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Yuan Liao</author>
        <description><![CDATA[Urban venues serve as arenas for social mixing, yet less is known about how public transit infrastructure shapes the geography of mixing at specific locations. This study examines how transit catchment diversity—the socioeconomic heterogeneity of populations reachable by public transit—associates with visitor diversity at points of interest (POIs) in nine Swedish and three US cities. Using mobile phone GPS traces and aggregated foot traffic data from 2024, we compute visitor diversity indices based on visitors' home-neighborhood birth-background composition and employ spatial regression models and geographically weighted regression (GWR). Transit catchment diversity positively predicts visitor diversity across nearly all cities, but this association is robust only in the largest metropolitan areas; in smaller cities, the coefficient attenuates to insignificance once geographic catchment composition, centrality, and venue density are controlled. Spatial spillovers in visitor diversity follow general geographic proximity rather than shared transit-stop connectivity, suggesting that the association operates through catchment population composition rather than station-level linkages. Transit–diversity hotspots occur not in already-diverse venues, but in lower-diversity POIs with lower commercial density, greater distance from transit in US cities, and greater centrality in Sweden. These patterns are consistent with transit-accessible population composition being associated with visitor diversity, particularly where alternative pathways to diverse co-presence are limited.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fdata.2026.1826953</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fdata.2026.1826953</link>
        <title><![CDATA[Inner layer security reinforcement for instant payment systems: a dual layer encryption-steganography evaluation in Brunei's digital payment context]]></title>
        <pubdate>2026-06-17T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Ampuan Shazani bin Ampuan Haji Sadikin</author><author>Heru Susanto</author>
        <description><![CDATA[IntroductionThe rapid adoption of real-time digital payment systems introduces cybersecurity risks that extend beyond technical vulnerabilities to include significant human and organizational factors.MethodsThis study evaluates a dual-layer data protection mechanism combining AES-128 encryption with spread spectrum audio steganography within Brunei's digital payment context. Stakeholder interviews with Cyber Security Brunei (CSB), National Digital Payments Network (NDPx), and a local bank were conducted alongside experimental testing using Stripe Sandbox.ResultsHuman factor issues accounted for 47% of identified cybersecurity concerns. Experimental results demonstrated that spread spectrum steganography achieved an 87.5% robustness rate across eight attack scenarios while maintaining near real-time performance with an average processing time of 568.82 ms and acceptable audio quality (PSNR 26.30 dB). Sandbox validation confirmed feasibility within realistic payment workflows.DiscussionThe findings support data-centric security as a compensating control in human-dominated threat environments and demonstrate the viability of combining encryption and steganography to reinforce instant payment security.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fdata.2026.1838191</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fdata.2026.1838191</link>
        <title><![CDATA[Measuring the impact of virtualization and containerization on the environment when using GPUs for processing the AI models]]></title>
        <pubdate>2026-06-16T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Safaa Hriez</author><author>Mohammad Haikal</author>
        <description><![CDATA[IntroductionThe rapid growth of artificial intelligence (AI) has significantly increased computing demand, intensifying the operational strain on the computing environment. While virtualization and containerization are established technologies for resource optimization, their comparative energy efficiency and environmental impact, particularly under GPU-accelerated AI workloads, are not well-quantified.MethodsThis study evaluates the energy consumption and environmental impact of virtualization and containerization technologies when using Graphics Processing Units (GPUs) for AI model execution. Employing a computer vision benchmark, the performance, GPU resource utilization, and power consumption were measured. The experiment involved training a DenseNet-121 model on the MNIST dataset within a VirtualBox virtual machine and a Docker container environment.ResultsThe analysis indicates that containerization consistently surpasses virtualization in energy efficiency. Specifically, Docker container configuration demonstrated an approximately 21.6% reduction in total energy consumption, and a corresponding reduction in carbon dioxide (CO2) emissions compared to a VirtualBox virtual machine. Furthermore, containerization exhibited lower average and peak GPU utilization and power consumption.DiscussionThese findings demonstrate that containerization offers a more energy-efficient and environmentally sustainable approach than VirtualBox virtualization for the specific GPU-enabled AI workload evaluated in this study. Statistical significance testing indicates that the observed performance differentials are significant, supporting the validity of the results within the experimental scope of this work.ConclusionImplementing containerization in this experimental setup may reduce energy consumption and environmental impact without compromising computational performance. Future studies should extend these analyses to larger neural network models, diverse AI workloads, and heterogeneous GPU platforms to enhance the generalizability of these findings beyond the current single-system experimental configuration.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fdata.2026.1835663</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fdata.2026.1835663</link>
        <title><![CDATA[Using artificial intelligence to improve governance and public services in Africa]]></title>
        <pubdate>2026-06-15T00:00:00Z</pubdate>
        <category>Mini Review</category>
        <author>David Mhlanga</author>
        <description><![CDATA[African governments continue to face persistent challenges in delivering efficient, transparent, and inclusive public services due to institutional constraints, rapid population growth, and fragmented administrative systems. At the same time, accelerating digital transformation and expanding data ecosystems create new opportunities for governance reform. This study examines the role of Artificial Intelligence (AI) in enhancing public sector performance across welfare targeting, healthcare delivery, tax administration, and urban governance. Drawing on a structured narrative literature review, the paper develops a conceptual framework that conceptualizes governance outcomes as a function of data availability, AI capability, institutional capacity, and human oversight. The findings suggest that AI can improve service delivery by enabling predictive decision-making, reducing administrative inefficiencies, and enhancing targeting accuracy. However, these benefits depend on the alignment between technological adoption and institutional readiness, as weak governance systems may amplify risks such as bias, exclusion, and accountability gaps. The study concludes that AI must be embedded in inclusive, context-sensitive governance strategies to support sustainable development outcomes in Africa.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fdata.2026.1752468</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fdata.2026.1752468</link>
        <title><![CDATA[Data field theory: a geometric framework for learning on Riemannian manifolds with synthetic validation and limitation analysis]]></title>
        <pubdate>2026-06-11T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Mohammadreza Nehzati</author>
        <description><![CDATA[IntroductionConventional machine learning treats learning as parameter optimization, lacking a first-principles framework for phenomena like criticality, generalization, and causal structure. We introduce Data Field Theory (DFT), a mathematical framework modelling learning as the evolution of a data field governed by stochastic partial differential equations on Riemannian manifolds. This work aims to validate DFT's core predictions in settings where its geometric assumptions hold, while honestly assessing its empirical limitations.MethodsWe formulate learning as a field φ:M×ℝ≥0→ℝk evolving on a spherical manifold. To test DFT, we implement a hierarchical classification task using synthetic data drawn from von Mises-Fisher distributions, ensuring match with the manifold geometry. We derive four key predictions: (1) critical exponents near concept formation, (2) a spectral robustness law linking Eigen gaps to out-of-distribution (OOD) error, (3) finite-speed causal propagation from hyperbolic regularization, and (4) approximate rotational equivariance via a Ward identity. We also conduct a preliminary real-data experiment projecting MNIST digits onto the sphere.ResultsSynthetic experiments validate all four predictions: (1) Correlation length diverges as ξ(t)~|t-tc|-ν with ν = 0.63 ± 0.04, accompanied by 1/f fluctuations; (2) OOD generalization error scales as ϵOOD∝mgap-2 (ρ = −0.78, p < 10−6); (3) Causal propagation speed ceff = 0.98 ± 0.03 (theory maximum cmax = 1.0) under hyperbolic regularization; and (4) Ward identity residual R = 0.0032 ± 0.0008 converging as R∝h1.02. However, on real-world MNIST-sphere data, DFT achieves only 15.7% accuracy versus 51.7% for k-NN, revealing critical limitations.DiscussionDFT successfully predicts emergent phenomena criticality, spectral robustness, bounded causality, and approximate equivariance under ideal geometric conditions, supporting its theoretical validity. The poor real-data performance highlights key gaps: the current framework lacks adaptive metric learning, noise robustness, and hierarchical feature extraction present in real images. These results establish DFT as a principled mathematical foundation for learning as field dynamics while clearly delineating necessary extensions for practical applicability.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fdata.2026.1736939</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fdata.2026.1736939</link>
        <title><![CDATA[Case count metric for comparative analysis of entity resolution results]]></title>
        <pubdate>2026-06-11T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>John R. Talburt</author><author>Muzakkiruddin Ahmed Mohammed</author><author>Mert Can Cakmak</author><author>Onais Khan Mohammed</author><author>Mahboob Khan Mohammed</author><author>Khizer Syed</author><author>Leon Claassens</author>
        <description><![CDATA[IntroductionEntity resolution (ER) systems often produce different clustering outcomes when applied to the same dataset, especially when parameters, algorithms, or system configurations change. However, in many real-world settings, the true linking structure is unknown, making traditional accuracy-based evaluation difficult.MethodsThis paper presents the Case Count Metric System (CCMS), a process and software system for comparing two cluster ER outcomes without requiring a truth set. CCMS classifies how each cluster from the first ER process is transformed by the second process into four mutually exclusive cases: unchanged, merged, partitioned, or overlapping.ResultsCCMS produces aggregate case counts, singleton summaries, and per-cluster transformation details to support diagnostic analysis. Example applications using synthetic demographic data and an industrial materials dataset show that CCMS can identify how clustering outcomes change under parameter adjustments and alternative ER systems.DiscussionCCMS provides a practical and interpretable method for comparing ER clustering results when labeled ground truth is unavailable. By distinguishing between over-linking, under-linking, and more complex cluster reorganizations, CCMS offers more actionable insight than single-value similarity measures and supports both research analysis and operational ER evaluation.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fdata.2026.1883246</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fdata.2026.1883246</link>
        <title><![CDATA[Correction: Explainable gradient convolutional vector fuzzy pattern analysis based on ensemble model for facial expression recognition]]></title>
        <pubdate>2026-06-10T00:00:00Z</pubdate>
        <category>Correction</category>
        <author>Lakshmi Sarvani Videla</author><author>Babu Reddy Mukamalla</author>
        <description></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fdata.2026.1825213</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fdata.2026.1825213</link>
        <title><![CDATA[When uncertainty guides learning: a highly effective approach to kidney disease classification in CT imaging]]></title>
        <pubdate>2026-06-09T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Muslima Akter</author><author>Fahmid Al Farid</author><author>Md Yousuf Ahmad</author><author>Md Azad Hossain Raju</author><author>Sowad Rahman</author><author>Jia Uddin</author><author>Hezerul Bin Abdul Karim</author>
        <description><![CDATA[The high cost of expert annotations significantly hinders the advancement of deep learning models for clinical medical imaging. This work introduces an efficient entropy-based active learning framework that achieves outstanding classification performance for renal abnormalities (Normal, Cyst, Stone, Tumor) in CT scans while requiring only a minimal amount of labeled data. The dataset comprises 12,446 CT slices split 70/15/15 into training (8,716), validation (1,865), and test (1,865) partitions via stratified sampling. Starting with only 200 randomly selected images and employing predictive entropy for uncertainty sampling on a pretrained ResNet-50 backbone, the proposed method attains 99.71% ± 0.25% mean test accuracy (95% CI: [99.30, 99.94]) across five independent runs after just six query cycles on the standard 12,446-image CT kidney dataset. Our method uses only 2,000 labeled training images, representing 22.9% of the 8,716-image training partition (a 77.1% reduction in required annotations relative to full supervision of the training set). This performance matches or exceeds prior fully supervised methods trained on the complete labeled training partition while demonstrating substantially improved sample efficiency, particularly in early annotation cycles where entropy-guided selection converges significantly faster than random sampling. Statistical testing across five repeated runs confirms that results are stable (Shapiro-Wilk p = 0.148). The framework exhibits exceptional sample efficiency as described by an empirically fitted power-law curve with a fitted exponent of 1.2, and empirically observed uncertainty decay with a rate of 0.92. These results offer both practical insights into annotation efficiency and substantial application value in the medical imaging domain.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fdata.2026.1785710</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fdata.2026.1785710</link>
        <title><![CDATA[Democratizing cloud data lake analytics: natural language access to Apache Iceberg via LLM agents]]></title>
        <pubdate>2026-06-08T00:00:00Z</pubdate>
        <category>Technology and Code</category>
        <author>Vipin Kataria</author><author>Nitin Kumar</author>
        <description><![CDATA[Business analysts and non-technical users need insights from enterprise data lakes but lack SQL expertise to query them directly. While large language models (LLMs) can translate natural language to SQL, existing text-to-SQL approaches face critical limitations: severe SQL injection vulnerabilities, inability to leverage data-lake-specific features like time-travel queries, and inconsistent metric definitions across organizations. We present the LangChain Iceberg Toolkit, enabling users to query Apache Iceberg data lakes through natural language conversations with LLM agents, no SQL knowledge required. Users ask questions in plain English (e.g., “What was revenue last quarter?”), and the system automatically: (1) interprets intent using LLMs, (2) selects appropriate tools from a YAML-based semantic layer mapping business terms to data structures, (3) executes queries through a hybrid architecture combining PyIceberg's type-safe API (for security) with DuckDB's SQL engine (for complex analytics), and (4) returns formatted answers with business context. Our evaluation demonstrates 100% success across 100 systematically designed queries leveraging semantic layer integration for consistent metric definitions. Critically, in direct comparison against a schema-aware text-to-SQL baseline on the same query set, our system achieves a 33 percentage-point accuracy improvement (100% vs. 67%) while reducing SQL injection attack success rate from the 99% reported in prior text-to-SQL research to 0% across both execution paths. End-to-end query latency averages 2.6 seconds on 15.1M records, with partition pruning eliminating 90%+ of scanned data files. The hybrid execution architecture prevents SQL injection vulnerabilities through type-safe query construction for simple queries and controlled, pre-validated SQL execution for complex analytics. Users receive data insights through conversational interfaces without writing SQL, understanding schemas, or knowing technical implementation details. We provide a production-ready, open-source implementation demonstrating practical viability for democratizing enterprise data access.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fdata.2026.1796969</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fdata.2026.1796969</link>
        <title><![CDATA[TCMB: cross-model multi-level cross-attention network with Taylor-based loss for multimodal fake news detection]]></title>
        <pubdate>2026-06-05T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Santosh Kumar Banbhrani</author>
        <description><![CDATA[IntroductionThe rapid spread of misinformation across social media platforms, websites, and online communication channels has made fake news detection a critical task in the digital era. Although various computational approaches have been developed to identify fake news, many existing methods suffer from limitations such as biased training datasets and high rates of false positives and false negatives. To address these challenges, this study proposes a Multimodal Cross Attention Network with Taylor-based Cross Entropy Mean Bias (MMCN_TCMB) model for detecting multimodal fake news.MethodsThe proposed approach utilizes multimodal inputs consisting of textual and visual content obtained from fake news datasets. The textual information in news posts is first tokenized using Bidirectional Encoder Representations from Transformers (BERT). Feature extraction is then performed using Word2Vec and Term Frequency–Inverse Gravity Moment (TF-IGM). Simultaneously, images associated with news posts undergo preprocessing through Contrast Limited Adaptive Histogram Equalization and Histogram Equalization (CLAHE-HE), followed by feature extraction using ResNet. The extracted textual and visual features are combined and processed through the MMCN framework. The learning mechanism of the network is enhanced using the Taylor-based Cross Entropy Mean Bias (TCMB) loss function to improve classification performance.ResultsExperimental results demonstrate that the proposed MMCN_TCMB model achieves superior performance in multimodal fake news detection. The model attains a recall of 97.988%, precision of 96.223%, F1-score of 97.098%, and overall accuracy of 97.436%, outperforming existing methods.DiscussionThe findings indicate that integrating multimodal feature extraction with cross-attention mechanisms and the TCMB loss function significantly enhances the reliability and accuracy of fake news detection. The proposed framework effectively captures both textual and visual inconsistencies, making it a promising approach for combating misinformation in modern digital platforms.The code is available on:https://github.com/banbhrani84/MMCN_TCMB-Fake-News-.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fdata.2026.1813265</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fdata.2026.1813265</link>
        <title><![CDATA[Interpretable intrusion detection for IoT: a CNN-BiLSTM permutation importance framework for deep feature selection]]></title>
        <pubdate>2026-05-22T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Ibrahim Al-Shibly</author><author>Llorenç Burgas</author><author>Joaquim Massana</author>
        <description><![CDATA[Industrial intrusion detection systems (IDS) in Industrial Internet of Things (IIoT) environments have to address the problem of handling multi-feature temporally correlated network traffic and dynamic changes in attack patterns. Traditional filter-based feature selection methods, like Mutual Information (MI), only consider individual feature performance and may not be effective in dealing with non-linear feature dependencies. This may degrade detection performance, especially in class-imbalanced problems. To mitigate such challenges, this paper proposes a deep feature selection (DFS) framework that utilizes a hybrid Convolutional Neural Network (CNN) and Bidirectional Long Short-Term Memory (BiLSTM) model. The proposed framework assesses the importance of native features using permutation importance. In the proposed framework, the CNN model detects local features in the data, whereas the BiLSTM model detects bidirectional temporal features in the data. The importance of features is computed by assessing the performance degradation of the model using time-aware perturbations on individual features. These identified features that are most relevant are then used to train lightweight traditional machine learning models like decision tree, K-nearest neighbor (KNN), logistic regression, naïve Bayes, and random forest. This makes it easy to deploy in resource-constrained IIoT environments. The approach is tested on the CIC IIoT 2025 dataset. From the experimental results, it is clear that the CNN-BiLSTM DFS framework improves recall and F1-score compared to other feature selection approaches like MI. This is especially true in imbalanced settings. The decoupling of feature selection from offline and edge-side inference provides a balance between detection accuracy, robustness, and deployability in real-world IIoT settings.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fdata.2026.1762571</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fdata.2026.1762571</link>
        <title><![CDATA[Definitional ambiguity in cognitive warfare: a critical and systematic conceptual review through ideal-type analysis]]></title>
        <pubdate>2026-05-15T00:00:00Z</pubdate>
        <category>Systematic Review</category>
        <author>Per-Erik Nilsson</author><author>Andreas Haga</author><author>Kristina Hellström</author>
        <description><![CDATA[Cognitive warfare is a relatively new concept in both military and academic discourse. The article's purpose is to advance conceptual clarity regarding cognitive warfare and to support future policy-oriented and academic research that strengthens the field's conceptual and methodological foundations, understood here as the broader domain of communication and defense studies concerned with informational and cognitive forms of contestation. This article examines how the notion is conceptualized within the emerging body of research, drawing on a systematic literature review. With support from LLM-assisted analysis, the study employs an exploratory methodology to identify both conceptual commonalities and points of divergence. The review indicates that cognitive warfare remains an underdeveloped research field, characterized by broad assumptions and limited scientific rigor. While the concept may represent a reframing of long-standing practices, it may also serve a political function by drawing renewed attention to forms of influence and conflict that have been overshadowed in recent decades. The article concludes by outlining avenues for future interdisciplinary research, emphasizing the need for conceptual clarity, empirical operationalization, and a more nuanced understanding of how adversaries themselves articulate and employ cognitive warfare.]]></description>
      </item>
      </channel>
    </rss>