Share your research data

This dataset accompanies the article “An investigation of a novel public health campaign to improve understanding of persistent pain” and contains anonymised quantitative data collected before and after a series of public Pain Science Education (PSE) events delivered across the Tees Valley region in 2023. The research hypothesis was that attending a Flippin’ Pain public health event would be associated with immediate improvements in pain‑related beliefs and behavioural intentions among both the general public and healthcare professionals (HCPs). The dataset includes responses to the Pain Beliefs Questionnaire (PBQ) organic subscale—the primary outcome measure—which assesses biomedical versus contemporary biopsychosocial beliefs about pain. As described in the article, the PBQ is “a validated eight‑item questionnaire with a higher score indicating stronger biomedical beliefs regarding pain.” Secondary outcome measures capture beliefs about the safety and benefits of physical activity, self‑rated ability to manage pain, and intentions regarding opioid use. Demographic variables include age, gender, ethnicity, education level, pain history, and physical activity levels. Data were gathered using paper questionnaires completed immediately before and after each event, with optional electronic follow‑up at 3, 6 and 12 months. Events consisted of a 1‑hour PSE presentation followed by a 30‑minute panel discussion. As noted in the article, the educational content “communicated key broad concepts which were assessed by the outcome measures used,” but was not designed to teach to specific questionnaire items. The dataset shows clear immediate post‑event shifts in pain beliefs. PBQ scores improved for both groups: the public demonstrated a mean change of –0.76 and HCPs –0.41, indicating movement away from structural‑pathology beliefs and toward evidence‑based conceptualisations of pain. Similar positive shifts were observed across all secondary outcomes, including increased perceived safety and benefit of physical activity and more guideline‑concordant intentions regarding opioid use. These findings support the interpretation that PSE‑based public health events can rapidly influence understanding and behavioural intentions related to persistent pain. Follow‑up response rates were low, so long‑term effects cannot be inferred from the dataset. However, the data provide a detailed snapshot of immediate cognitive and behavioural intention changes following a real‑world public health intervention. Researchers may use the dataset to explore associations between demographic factors, pain history, and responsiveness to PSE; to replicate or extend analyses; or to inform future intervention design and evaluation. I can confirm that no personally identifiable data are included and that informed consent has been obtained to take part in this study and for the non-person identifiable data to be made publicly available on this repository.

Microbiome sequencing

This is the minimal dataset for applying MT-CNN U-Net to reconstruct the planar foot pressure from only 10 sparse sensors.

Supplemental material including PubMed search strings and PRISMA Flow Diagram

Liste des centres d'examen civique et TCF (test de connaissance du français) utilisés dans la procédure de naturalisation française : nom, adresse, code postal, ville, coordonnées GPS et site d'inscription. Publié par France Naturalisation Infos à partir du jeu de données sur data.gouv.fr (Licence Ouverte 2.0). Données non officielles : vérifiez les informations auprès de chaque centre. Site : https://france-naturalisation.net

This dataset accompanies the manuscript “Unified Parameterization and Flow-Regulation Mechanisms of Transonic Compressor Blade Profiles Based on an Exponential Camber Line.” It provides the geometric parameters, camber line coefficients, and surface coordinates used in the assessment of parameter-guided compressor profile modification. The dataset covers five profile categories: large-turning C-type, midspan C-type, tip C-type, J-type, and S-type pre-compression profiles. Each category includes an initial profile (R) and a modified profile (O) for both Set I and Set II, giving 20 geometries in total. In Set I, the four camber line shape parameters are adjusted while the inlet geometric angle and overall turning remain fixed. Set II additionally varies the inlet geometric angle and overall turning. Thickness parameters are held fixed within each initial–modified profile comparison. The parameter and coefficient table identifies the profile category, set, and initial or modified status of each geometry. Coordinate files provide the pressure-side and suction-side contours. Parameter definitions, units, coordinate conventions, and file naming are documented in the accompanying README file. These data support reconstruction of the parameterized geometries and comparison of the initial and modified profiles.

Core method implementation for the manuscript "Joint evidential-state analysis for unfamiliar-traffic diagnosis in Internet of Things intrusion detection". This archive contains the evidential neural network, the expected Dirichlet cross-entropy objective with annealed KL regularization, the two-coordinate evidence-state transformation (evidence insufficiency and normalized conditional-allocation entropy), known-reference fitting with the Ledoit-Wolf covariance estimator, and the full and matched-diagonal Mahalanobis scores, together with the model configuration reported in the paper and synthetic unit tests. Scope: this is a core-method release. It does not include raw traffic or packet captures, per-record processed inputs, trained checkpoints, dataset-specific preprocessing and bundle builders, or the plotting and table-production scripts. Source datasets must be obtained from their original providers under their own terms; aggregate numerical summaries supporting the reported results are provided with the manuscript's Supplementary Information. Tested with Python 3.10+; run python -m unittest discover -s tests -v.

This dataset contains raw VOC sensor-response data collected from naturally ripened (NRB) and artificially ripened bananas (ARB) over a 14-day monitoring period. Artificial ripening was induced using a 1.2 g/L ethephon-water solution for 5 minutes, while the remaining bananas were allowed to ripen naturally. A five-sensor MOS array consisting of TGS2602, TGS2603, TGS2610, TGS2612, and TGS2620 was connected to an NI myRIO and LabVIEW for automated data acquisition. VOCs were allowed to accumulate for 20 minutes, followed by 5 minutes of sensor-response measurement at 1-second intervals. The sensing chamber was subsequently cleared for 5 minutes. The dataset is systematically organized into two primary directories: “Artificially Ripened Banana Sensor-array Response” and “Naturally Ripened Banana Sensor-array Response.” Each directory contains 14 day-wise folders, designated “Sensor Data Day 1” through “Sensor Data Day 14.” The raw sensor responses are provided in CSV format without steady-state selection, averaging, or machine-learning preprocessing, allowing independent preprocessing and analysis. The dataset structure is as follows: VOC Dataset/ - Artificially Ripened Banana Sensor-array Response/ - Sensor Data Day 1–Day 14 - Naturally Ripened Banana Sensor-array Response/ - Sensor Data Day 1–Day 14 The complete dataset contains 1009 samples of raw sensor-array measurements.

Stocking density is usually evaluated through growth, which may not reveal its effects on fish welfare. Juvenile Nile tilapia from a sex-reversed and a non-reversed batch were stocked at 100, 200, 300, and 400 fish/m³ in a recirculating system for 56 days. Feeding behavior was recorded in 320 videos, and blood glucose, the hematological profile, and fillet oxidative markers were measured.

Supplementary materials for “Culture and Consumer Responses to Food Experiences: A Systematic Review”. The deposit contains S1, the study-level coding matrix for 122 included records and eligibility reassessment of 14 excluded records; S2, the MMAT 2018 appraisal of 98 primary empirical studies, with 24 non-primary records marked as not applicable; S3, the completed PRISMA 2020 checklist; S4, database-specific search strategies and documented retrieval counts; and S5, the cross-study evidence synthesis. The search coverage was updated to 31 August 2026. Historical stage-specific counts not retained in the review documentation have not been reconstructed.