Find research datasets worth reusing
Search datasets from major research repositories and use ShareScore to quickly assess how well each record supports discovery, access, and reuse.
832
datasets available to search
ShareScore release 0.9.0
Dataset results
832 results for “automatic”
Automatic affective reactions to physical effort
<p>Dataset for the study titled "automatic affective responses during physical effort: a virtual reality study".</p> <p>This dataset includes:</p> <p><strong>1) A codebook (including the name of the main variables)</strong></p> <p>--> "code_book_affect_effort.xlsx"</p> <p><strong>2) Behavioral data (raw)</strong></p> <p>--> in the folder "data_ps_VR". </p> <p>The raw data are added for transparency, but are not necessary to run the models. </p> <p><strong>3) Self-reported data (raw)</strong></p> <p>--> "20220112_VR_expe.xlsx"</p> <p>--> "20220112_VR_pilot.xlsx"</p> <p>The raw data are added for transparency, but are not necessary to run the models. </p> <p><strong>4) clean data ready to used for the statistical analyses</strong></p> <p>--> "data_VR_all_clean.csv".</p> <p>This clean data are produced by the R script. These data included the self-reported and the behavioral measures. </p> <p><strong>5) R script for the data management (i.e., from the raw data to data ready to be analyzed)</strong></p> <p>--> "data_management_effort.R" to create the dataset (return the file: "data_VR_all_clean.RData")</p> <p><strong>6) R script for the models tested</strong></p> <p>--> "Data_mixed_effects_models.R" for the models tested in the paper</p> <p><strong>7) The video of the experiment</strong></p>
Automatic weather station data from the debris-covered Kennicott Glacier, Alaska (May-Aug 2019)
<p>Data from an automatic weather station installed on the debris-covered<strong> Kennicott Glacier</strong> in the Wrangell St. Elias National Park, Alaska, USA, spanning most of the 2019 ablation season (28 May - 22 August 2019).</p> <p>Brief description of the uploaded datasets:</p> <ul> <li><em><strong>KEN_2019_60.csv</strong></em>: Hourly aggregated measurements of air temperature, relative humidity, wind speed and direction, atmospheric pressure, incoming and reflected shortwave radiation, incoming and outgoing longwave radiation and precipitation</li> <li><em><strong>KEN_2019_AWS_Metadata.pdf</strong></em>: Metedata file describing the station location, sensor setup and measurements</li> <li><em><strong>KEN_2019_debris.csv</strong></em>: Debris properties assumed to be potentially applicable for this site (not measured but taken from literature)</li> <li><strong><em>KEN_2019_VAL_debristemperature.csv</em></strong>: Hourly time series of temperature measurements at known depths within the debris cover close to the station location</li> <li><strong><em>KEN_2019_VAL_subdebrismelt.csv</em></strong>: Monthly stake measurements of surface height change (sub-debris ice melt) close to the station location spanning the ablation season</li> </ul> <p>We recorded additional meteorological data at the weather station (e.g. vertical profiles of wind speed/direction, relative humidity and temperature; 3D ultrasonic anemometer) as well as subdebris melt, dGPS-surface change and temperature profiles at additional transects for the same period. Please contact the authors in case of interest in any additional data.</p>
Automatic Number Plate Recognition Data from Bristol, England for Apr-Jun 2019, 2020 and 2021
<p>These files contain data derived from Automatic Number Plate Recognition (ANPR) data collected by Bristol City Council as part of their continuous traffic monitoring work. The data was used in a project examining the impact of Covid-19 on travel behaviour in Bristol and the implications for air quality. </p> <p>Number plate data cannot be shared, therefore the files contain anonymised vehicle identifiers. Also, specific information about the locations of the ANPR cameras cannot be provided under the data sharing agreement we have with the data owners. Figures 4 and 19 in the project report (see uploaded pdf file) indicate the approximate locations of the ANPR cameras. </p> <p>The following files have been uploaded: </p> <p>- a metadata file containing descriptions of variables in the vehicle and trip files </p> <p>- a copy of the report which came out of this project. </p> <p>- BCCANPR21_REG_descrip_withID_NOVRM.csv: This file includes vehicle information for all vehicles observed in the trip data. The anonymised identification numbers can be used to merge this data with the trip files. </p> <p>- BCCANPR21_2019AprMayJun_REG_idv2.csv: Three files of this format contain trip data. Each file contains trips in April to June of the year specified in the file name. The number plate data was processed to chain together observations that were part of the same trip and this file only contains the location of the first and last ANPR camera at which this trip was recorded. </p> <p>This project was funded by the TRANSITION Clean Air Network through the first round of its Discovery & Innovation Fund in 2021. TRANSITION is a UK-wide network, led by the University of Birmingham in collaboration with nine universities and over 20 cross-sector partners, aiming to optimise the air quality and health outcomes of transport decarbonisation. The network (NERC ref. NE/V002449/1) is itself funded by UK Research & Innovation through its Clean Air Strategic Priorities Fund, administered by the Natural Environment Research Council. </p> <p>The number plate data used in this research was provided by the Bristol Traffic Control Service at Bristol City Council. The vehicle data was obtained through the DVLA’s Vehicle Enquiry Service API. </p>
InsectSet32: Dataset for automatic acoustic identification of insects (Orthoptera and Cicadidae)
<p>This dataset contains recordings of 32 sound producing insect species with a total 335 files and a length of 57 minutes. The dataset was compiled for training neural networks to automatically identify insect species while comparing adaptive, waveform-based frontends to conventional mel-spectrogram frontends for audio feature extraction. This work was <a href="https://doi.org/10.1371/journal.pcbi.1011541">published</a> in PLOS Computational Biology and this dataset can be used to replicate the results, as well as other uses. The scripts for audio processing and the machine learning implementations are published on <a href="https://github.com/mariusfaiss/InsectSet32-Adaptive-Representations-of-Sound-for-Automatic-Insect-Recognition">Github</a>.</p> <p>The recordings are split into two datasets. Roughly half of the recordings (147) are of nine species belonging to the order Orthoptera. These recordings stem from a dataset that was originally compiled by <a href="https://orcid.org/0000-0002-8929-2737">Baudewijn Odé</a> (unpublished). </p> <p>The remaining recordings (188) are of 23 species in the family Cicadidae. These recordings were selected from the Global Cicada Sound Collection hosted on <a href="https://bio.acousti.ca/">Bioacoustica</a> (<a href="https://doi.org/10.1093/database/bav054">doi.org/10.1093/database/bav054</a>), including recordings published in <a href="https://doi.org/10.3897/BDJ.3.e5792">doi.org/10.3897/BDJ.3.e5792</a> & <a href="https://doi.org/10.11646/zootaxa.4340.1">doi.org/10.11646/zootaxa.4340.1</a>. Many recordings from this collection included speech annotations in the beginning of the recordings, therefore the last ten seconds of audio were extracted and used in this dataset. </p> <p>All files were manually inspected and files with strong noise interference or with sounds of multiple species were removed. Between species, the number of files ranges from four to 22 files and the length from 40 seconds to almost nine minutes of audio material for a single species. The files range in length from less than one second to several minutes. All original files were available with sample rates of at least 44.1 kHz or higher but were resampled to 44.1 kHz mono WAV files for consistency. The annotation files contain information for each recording, including the file name, species name and identifier, as well as the data subset they were included in for training the neural network (training, test, validation).</p>
Dataset of Automatically Orchestrable GitHub Projects
<p>This dataset accompanies the submission "Generating representative, live network traffic out of millions of code repositories" at HotNets'22: The 21st ACM Workshop on Hot Topics in Networks.</p> <p>Please see the files:<br> - `list_of_github_repositories.txt` for a list of GitHub repositories that we found containing a `docker-compose*.yml` file<br> - `list_of_executed_repositories.csv` for more detailed information on the success of capturing traffic with specific orchestration files found in ~67% of the repositories<br> <br> If you use our dataset, please cite our work as follows:</p> <blockquote> <p>Tobias Bühler, Roland Schmid, Sandro Lutz, and Laurent Vanbever.<br> 2022. Generating representative, live network traffic out of millions<br> of code repositories. In The 21st ACM Workshop on Hot Topics<br> in Networks (HotNets ’22), November 14–15, 2022, Austin, TX,<br> USA. ACM, New York, NY, USA, 7 pages. https://doi.org/10.1145/<br> 3563766.3564084</p> </blockquote>
An Automatically Transcribed Piano Performer Dataset
<p>Data-driven approaches for performers' style analysis need large corpora of music performances to derive expressive performance parameters. One good reason for Deep Neural Networks (DNNs) not being used for performer identification is the lack of large-scale datasets with overlapping performances by different performers. Hence, to bridge this gap, we created a score-aligned automatically transcribed performer dataset. There are a total of 474 performances in the dataset, played by 6 pianists, spanning 35 movements by 2 composers for a total of 474 Western classical piano recordings in MIDI format. Every single MIDI file that's included in this collection was derived from a piano transcription of an already-existing audio recording of a piano performance. Score files in musicXML format and the alignment results are also included in the dataset.</p>
Conversational Networks For Automatic Online Moderation
<p><strong>Description. </strong>This repository contains several datasets of conversational networks, extracted from the chat messages exchanged by players of the <a href="https://www.spaceorigin.fr/">SpaceOrigin</a> MMORPG. Each graph represents a specific conversation, and belongs to one of two classes: <em>Abusive</em> (1) or <em>Non-abusive</em> (0). Vertices represent users, and edges represent the fact that the connected users exchanged message during the considered time period. Edges are weighted and directed: weights represent the intensity of the message exchanges, and directions represent who sent messages to whom.</p> <p>We provide two types of graphs: unsigned and signed. Unsigned graphs were extracted using the method described in paper [1], below. Version 1.0 of this dataset contain only a part of the conversations, subsampled to get balanced classes. Version 1.1 is extended to contain all available conversations, and there are much more <em>Non-abusive</em> than <em>Abusive</em> conversations. Signed graphs were extracted later, using the method described in publication [9] below. Each edge is described by an additional sign, that indicates the polarity of the messages exchanged by two users; friendly (positive) vs. hostile (negative). </p> <p>These datasets were used to train a classifier into automatically recognizing abusive messages. See the below papers for more details. The repository also contains some figures that appear in these papers.</p> <p><strong>Publications. </strong>The following papers used the unsigned version of the conversational networks. The extraction method is described in paper [1].</p> <ul> <li>[1] É. Papégnies, V. Labatut, R. Dufour & G. Linarès, “Conversational Networks for Automatic Online Moderation,” <em>IEEE Transactions on Computational Social Systems </em>6(1):38–55, 2019. ⟨<a href="https://hal.archives-ouvertes.fr/hal-01999546">hal-01999546</a>⟩ DOI: <a href="https://doi.org/10.1109/TCSS.2018.2887240">10.1109/tcss.2018.2887240</a></li> <li>[2] É. Papegnies, R. Dufour, V. Labatut & G. Linarès. “Détection de messages abusifs au moyen de réseaux conversationnels,” in <em>8ème Conférence sur les modèles et l'analyse de réseaux : approches mathématiques et informatiques</em> (MARAMI), 2017. ⟨<a href="https://hal.science/hal-01614279">hal-01614279</a>⟩</li> <li>[3] É. Papegnies, V. Labatut, R. Dufour, & G. Linares. “Graph-based Features for Automatic Online Abuse Detection,” in <em>International Conference on Statistical Language and Speech Processing</em> (SLSP), Springer, Lecture Notes in Computer Science 10583:70-81, 2017. ⟨<a href="https://hal.science/hal-01571639/">hal-01571639</a>⟩ DOI: <a href="http://doi.org/10.1007/978-3-319-68456-7_6">10.1007/978-3-319-68456-7_6</a></li> <li>[4] N. Cécillon. “Exploration de descripteurs de plongements de graphes pour la détection de messages abusifs,” MSc Thesis, <em>Université d'Avignon, </em>2019. ⟨<a href="https://dumas.ccsd.cnrs.fr/dumas-04073337">dumas-04073337</a>⟩</li> <li>[5] N. Cécillon, V. Labatut, R. Dufour, and G. Linarès, “Abusive Language Detection in Online Conversations by Combining Content- and Graph-based Features,” in <em>International Workshop on Modeling and Mining Socia-Media Driven Complex Networks</em>, Frontiers in Big Data 2:8, 2019. ⟨<a href="https://hal.archives-ouvertes.fr/hal-02130205">hal-02130205</a>⟩ DOI: <a href="http://doi.org/10.3389/fdata.2019.00008">10.3389/fdata.2019.00008</a></li> <li>[6] N. Cécillon, V. Labatut, R. Dufour, & G. Linarès. “Tuning Graph2vec with Node Labels for Abuse Detection in Online Conversations,” in <em>11ème Conférence sur les modèles et l'analyse de réseaux : approches mathématiques et</em> informatiques (MARAMI), 2020. <a href="https://hal.science/hal-02993571" target="_blank" rel="noopener">⟨hal-02993571⟩</a> <a href="http://ceur-ws.org/Vol-2750/paper8.pdf">Official Page</a></li> <li>[7] N. Cécillon, V. Labatut, R. Dufour & G. Linarès. “Graph embeddings for Abusive Language Detection,” <em>Springer Nature Computer Science</em> 2:37,<em> </em>2021<em>. </em>⟨<a href="https://hal.science/hal-03042171/">hal-03042171</a>⟩ DOI: <a href="http://doi.org/10.1007/s42979-020-00413-7">10.1007/s42979-020-00413-7</a></li> <li>[8] N. Cécillon, R. Dufour & V. Labatut. “Approche multimodale par plongements de texte et de graphes pour la détection de messages abusifs,” <em>Traitement Automatique des Langues</em><em> </em>62:13-38, 2021. ⟨<a href="https://hal.science/hal-03527016">hal-03527016</a>⟩ <a href="https://aclanthology.org/2021.tal-2.2/">Official Page</a></li> </ul> <p>The following publications use the signed version of the graphs. The modified extraction method is described in publication [9].</p> <ul> <li>[9] N. Cécillon. “Combining Graph and Text to Model Conversations: An Application to Online Abuse Detection,” PhD Thesis, <em>Université d'Avignon, </em>2024. ⟨<a href="https://theses.hal.science/tel-04441308">tel-04441308</a>⟩</li> </ul> <p><strong>Funding. </strong>Part of this work was funded by a grant from the <em>Provence-Alpes-Côte-d'Azur</em> region (PACA, France) and the <em>Nectar de Code</em> company.</p> <p><strong>Citation. </strong>If you use this dataset, please cite paper [1] for the unsigned networks:</p> <p><br><code>@Article{Papegnies2019,</code><br><code> author = {Papegnies, Étienne and Labatut, Vincent and Dufour, Richard and Linarès, Georges},</code><br><code> title = {Conversational Networks for Automatic Online Moderation},</code><br><code> journal = {IEEE Transactions on Computational Social Systems},</code><br><code> year = {2019},</code><br><code> volume = {6},</code><br><code> number = {1},</code><br><code> pages = {38-55},</code><br><code> doi = {10.1109/TCSS.2018.2887240},</code><br><code>}</code></p> <p>and [9] for the signed ones:</p> <p><br><code>@PhdThesis{Cecillon2024,</code><br><code> author = {Cécillon, Noé},</code><br><code> title = {Combining Graph and Text to Model Conversations: An Application to Online Abuse Detection},</code><br><code> school = {Université d'Avignon},</code><br><code> year = {2024},</code><br><code> type = {PhD Thesis},</code><br><code> address = {Avignon, FR},</code><br><code> url = {https://theses.fr/2024AVIG0100},</code><br><code>}</code></p>
Multifunctional Polymer Composites for Automatable Induction Heating with Subsequent Temperature Verification
<p>This data upload contains the metadata and datasets underlying the manuscript: "Multifunctional Polymer Composites for Automatable Induction Heating with Subsequent Temperature Verification".</p> <p>A description of the uploaded data is found in the README.txt.</p>
An Automatic Neuroimaging Infrastructure For Synthesis and Analysis of Structural MRI Data
<p>We have designed, implemented and distributed a fully automatic neuroimaging infrastructure for the synthesis and analysis of structural magnetic resonance imaging (MRI) data. The framework provides a concrete environment for the quantitative validation of various methods for the analysis of brain asymmetries, for comparisons of methods and measures of brain shape asymmetry, and possibly for clarifying contradicting neuroimaging findings of brain lateralizations.</p> <p>See <a href="https://sites.google.com/site/brainmorphorg/home">https://sites.google.com/site/brainmorphorg/home </a></p> <p>and </p> <p>A. Pepe, I. Dinov, and J. Tohka . An Automatic Framework for Quantitative Validation of Voxel Based Morphometry Measures of Anatomical Brain Asymmetry. <a href="http://dx.doi.org/10.1016/j.neuroimage.2014.06.029">NeuroImage , 100: 444 - 459, 2014</a><a href="https://doi.org/10.1016/j.neuroimage.2014.06.029"> </a></p> <p>for more information. </p>
A Semi-Automatic Classification Approach for River Shape Extraction from Sentinel-2 Imagery
<p>To extract the river from satellite imagery, at first we have to classify the waterbody from satellite imagery. Then we will differentiate the river from waterbody. We have used three different techniques to classify the waterbody from satellite imagery. At first, pixel based iso-cluster unsupervised classification was used to classify the waterbody in Sentinel 2 imagery. We excluded the supervised classification in decided methodology as we are interested about automatic process of river extraction. Then we have used Segment mean shift classification tool as image segmentation techniques. Indices based NDWI (Normalize difference water indices) classification was also used in this research.</p>
Evaluation Data of the Implementation of the Approach for Automatic Test Generation for Information-Flow Properties
<p>This data set contains the programs for which the automatic test generation approach of the KeY theorem prover was used to automatically generate noninterference tests.</p> <p>The approach is described in <a href="http://dx.doi.org/10.1145/3297280.3297500 ">http://dx.doi.org/10.1145/3297280.3297500 </a></p> <p>DATA<br> ---------<br> The data folder contains the secure and insecure programs which were evaluated and the tests which were generated for them.</p> <p>Each program is in the folder "program" and is written in Java and specified in an extended version of the JML specification language. Check out <a href="http://dx.doi.org/10.5445/IR/1000046878">http://dx.doi.org/10.5445/IR/1000046878</a> for a reference on the used specification language.</p> <p>For each example we provide the tests that were generated. For the insecure examples we provide the tests generated with each of the two options of our approach. The tests generated with the option for searching for counterexamples is in the folder "WithPost" of each insecure example.</p> <p> </p>
Automatic TEI encoding of manuscripts catalogues with GROBID-Dictionaries
<p>Manuscript Sales Catalogues (MSC) are highly important for authenticating documents and studying the reception of authors. Their regular publication throughout Europe since the beginning of the 19th c. has consequently raised the interest around scaling up the means for automatically structuring their contents. </p> <p>Following successful first encoding tests with <em>GROBID-Dictionaries</em> on a single MSC collection, we aim in this paper to present the results of more advanced tests of the system’s capacity to handle a larger corpus with MSC of different dealers, and therefore multiple layouts. Four different types of catalogues published between the middle of the 19th c. and the beginning of the 20th c. have been tested.</p>
Improving Automatic Melanoma Diagnosis using Deep Learn-ing-based Segmentation of Irregular Networks
<p>Irregular masks dataset created on a subset of the ISIC19 training dataset. All annotations are for melanoma lesions. The filename indicates the ISIC19 image id along with suffix indicating annotator and/or verifier. This dataset was used in the publication "Improving Automatic Melanoma Diagnosis using Deep Learn-ing-based Segmentation of Irregular Networks" to be submitted to the Cancers Journal.</p> <p>Please cite the corresponding article (to be published) if data is used in your work.</p> <p>The references for the ISIC19 dataset that this is built on is given below.</p> <blockquote> <p>BCN_20000 Dataset: (c) Department of Dermatology, Hospital Clínic de Barcelona</p> <p>HAM10000 Dataset: (c) by ViDIR Group, Department of Dermatology, Medical University of Vienna; <a href="https://doi.org/10.1038/sdata.2018.161">https://doi.org/10.1038/sdata.2018.161</a></p> <p>MSK Dataset: (c) Anonymous; <a href="https://arxiv.org/abs/1710.05006">https://arxiv.org/abs/1710.05006</a>; <a href="https://arxiv.org/abs/1902.03368">https://arxiv.org/abs/1902.03368</a></p> </blockquote>
Automatically Reproducing Timing-Dependent Flaky-Test Failures
<p>This artifact contains the source code for FlakeRake, a tool for automatically reproducing timing-dependent flaky-test failures. It also includes raw and processed results produced in the evaluation of FlakeRake</p> <p> </p> <p>Contents:</p> <p> </p> <p>Timing-related APIs that FlakeRake considers adding sleeps at: timing-related-apis</p> <p>Anonymized code for FlakeRake (not runnable in its anonymized state, but included for reference; we will publicly release the non-anonymized code under an open source license pending double-blind review): flakerake.tgz</p> <p>Failure messages extracted from the FlakeFlagger dataset: 10k_reruns_failures_by_test.csv.gz </p> <p>Output from running isolated reruns on each flaky test in the FlakeFlager dataset: 10k_isolated_reruns_all_results.csv.gz (all test results summarized into a CSV), 10k_isolated_reruns_failures_by_test.csv.gz (CSV including just test failures, including failure messages), 10k_isolated_reruns_raw_results.tgz (includes all raw results from reruns, including the XML files output by maven)</p> <p>Output from running the FlakeFlagger replication study (non-isolated 10k reruns):flakeFlaggerReplResults.csv.gz (all test results summarized into a CSV), 10k_reruns_failures_by_test.csv.gz (CSV including just failures, including failure messages), flakeFlaggerRepl_raw_results.tgz (includes all raw results from reruns, including the XML files output by maven - this file is markedly larger than the 10k isolated reruns results because we ran *all* tests in this experiment, whereas the 10k isolated rerun experiment only re-ran the tests that were known to be flaky from the FlakeFlagger dataset).</p> <p>Output from running FlakeRake on each flaky test in the FlakeFlagger dataset:</p> <p>For bisection mode: results-bis.tgz</p> <p>For one-by-one mode: results-obo.tgz</p> <p>Scripts used to execute FlakeRake using an HPC cluster: execution-scripts.tgz<br> Scripts used to execute rerun experiments using an HPC cluster: flakeFlaggerReplScripts.tgz<br> Scripts used to parse the "raw" maven test result XML files in this artifact into the CSV files contained in this artifact: parseSurefireXMLs.tgz </p> <p>Output from running FlakeRake in “reproduction” mode, attempting to reproduce each of the failures that matched the FlakeFlagger dataset (collected for bisection mode only): results-repro-bis.tgz</p> <p>Analysis of timing-dependent API calls in the failure inducing configurations that matched FlakeFlagger failures: bis-sleepyline.cause-to-matched-fail-configs-found.csv</p>
An Open-Source Automatic Survey of Green Roofs in London using Segmentation of Aerial Imagery: Dataset
<p>This archive contains code and data to go with the paper <em>*An Open-Source Automatic Survey of Green Roofs in London using Segmentation of Aerial Imagery*</em>.</p> <p> </p> <p>This archive contains geospatial data, as well as the code used to generate the geospatial data.</p> <p>The geospatial data consists of georeferenced polygons identifying areas which are covered by green roofs in London (GBR) generated from 2019 aerial imagery.</p> <p>The data is described in detail in the manuscript <em>*An Open-Source Automatic Survey of Green Roofs in London using Segmentation of Aerial Imagery*</em>. See abstract below.</p> <p> </p> <p>GeoJSON format:</p> <p>GeoJSON is a format for encoding geospatial data, see https://geojson.org/.</p> <p>GeoJSON can be read using GIS programs including ArcGIS, QGIS, OGR.</p> <p> </p> <p>Contents:</p> <p>`geospatial_data/buffered_polygons_2021.zip` a zip archive containing a geojson file. It is the estimated locations of green roofs in London in 2021 and is the main result, which can be opened in any GIS program after being unzipped.</p> <p>`geospatial_data/buffered_polygons_2019.zip` a zip archive containing a geojson file. It is the estimated locations of green roofs in London in 2019 and is a secondary result, which can be opened in any GIS program after being unzipped. The predictions were made with the same model as the 2021 results.</p> <p>`geospatial_data/labelled_area.zip` a zip archive containing a geojson file. Identifies the area which was hand-labelled.</p> <p>`geospatial_data/manual_2021.zip` a zip archive containing a geojson file. Manually labelled green roof from 2021 imagery.</p> <p>`geospatial_data/manual_2019.zip` a zip archive containing a geojson file. Manually labelled green roof from 2019 imagery.</p> <p>`segmentation_code` contains the code used to produce the segmentation from the aerial imagery.</p> <p>`analysis_code` contains the code used to produce the plots and tables for the paper.</p> <p> </p> <p>Imagery availability:</p> <p>Unfortunately the aerial imagery and building footprint data cannot be shared directly, as you will require the proper license. Both can be found at [Digimap](https://digimap.edina.ac.uk) provided your institution has the license.</p> <p> </p> <p>Abstract:</p> <p>Green roofs can mitigate heat, increase biodiversity, and attenuate storm water, giving some of the benefits of natural vegetation in an urban context where ground space is scarce. To guide the design of more sustainable and climate resilient buildings and neighbourhoods, there is a need to assess the existing status of green roof coverage and explore the potential for future implementation. Therefore, accurate information on the prevalence and characteristics of existing green roofs is needed, but this information is currently lacking. Segmentation algorithms have been used widely to identify buildings and land cover in aerial imagery. Using a machine-learning algorithm based on U-Net to segment aerial imagery, we surveyed the area and coverage of green roofs in London, producing a geospatial dataset \cite[]{simpson_charles_2022_6861929}. We estimate that there was 0.23 km^2 of green roof in the Central Activities Zone (CAZ) of London, (1.07 km^2) in Inner London, and (1.89 km^2) in Greater London in the year 2021. This corresponds to 2.0% of the total building footprint area in the CAZ, and 1.3% in Inner London. There is a relatively higher concentration of green roofs in the City of London, covering 3.9% of the total building footprint area. Test set accuracy was 0.99, with an f-score of 0.58. When tested against imagery and labels from a different year (2019), the model performed just as well as a model trained on the imagery and labels from that year, showing that the model generalised well between different imagery. We improve on previous studies by including more negative examples in the training data, and by requiring coincidence between vector building footprints and green roof patches. We experimented with different data augmentation methods, and found a small improvement in performance when applying random elastic deformations, colour shifts, gamma adjustments, and rotations to the imagery. The survey covers 1558 km^2 of Greater London, making this the largest open automatic survey of green roofs in any city. The geospatial dataset is at the single-building level, providing a higher level of detail over the larger area compared to what was already available. This dataset will enable future work exploring the potential of green roofs in London and on urban climate modelling.</p>
A software for automatic identification of oyster species
<p>The files includes all data and final analysis of the work done in CS8 - oysters, task 8.2, CSTP8.2.2_A new software for automatic identification of oyster species. This includes the data management descriptor document (DataSheet_oyster_image_classification.docx), images used (oyster_classification_images.zip), the code developed (oyster_classification_code_package.zip) and different models evaluated (oyster_classification_models.zip), the genetics data produced (oyster_classification_biometrics and PCR.xlsx) and the project report (C639_ostronklassificering.pdf). The content of the files is described briefely below. The data is used in deliverables D1.2, D1.4, D1.5 and D1.6 in the AquaVitae project.</p> <p>oyster_classification_images.zip</p> <p>The data set contains the images used for training the classification models that are capable of classifying images of oysters as either Ostrea edulis or Magallana gigas. The images are sorted in folders named “train” (training data) and “validation” (validation data) with both folders containing sub-folders called “mg” (images of Magallana gigas) and “oe” (images of Ostrea edulis).</p> <p>oyster_classification_code_package.zip</p> <p>The data set contains the code for training a neural network for classifying oyster species based on images. The code also includes localization of oyster within an image and inference of the classification along with the trained models.</p> <p>oyster_classification_models.zip</p> <p>The data set contains the trained classification models that are capable of classifying images of oysters as either Ostrea edulis or Magallana gigas.</p> <p>oyster_classification_biometrics and PCR.xlsx</p> <p>The data set contains biometric information for a subset of 240 Ostrea edulis, 240 Magallana gigas and 204 oysters of unsure species denotation sampled as a start pool for the image analysis project and for genetic evaluation of species belonging.</p>
ITTV - A Dataset of Italian Television for Automatic Genre Classification
<p>ITTV is a publicly available dataset of Italian TV programs introduced in </p> <blockquote> <p>Alessandro Ilic Mezza, Paolo Sani, and Augusto Sarti, "Automatic TV Genre Classification Based on Visually-Conditioned Deep Audio Features," in 2023 31st European Signal Processing Conference (EUSIPCO), 2023.</p> </blockquote> <p>ITTV consists of 2625 manually annotated YouTube videos, totaling over 670 hours. Each clip is assigned one of seven classes:</p> <ul> <li>Cartoons</li> <li>Commercials</li> <li>Football</li> <li>Music</li> <li>News</li> <li>Talk Shows</li> <li>Weather Forecast</li> </ul> <p>ITTV genre taxonomy is similar to that of the well-known RAI dataset described in</p> <blockquote> <p>Maurizio Montagnuolo and Alberto Messina, "Parallel neural networks for multimodal video genre classification,” Multimedia Tools and Applications, vol. 41, no. 1, pp. 125–159, 2009.</p> </blockquote> <p>The dataset contains genre annotations and metadata in CSV format. Please note that audio data is not provided.</p> <p>We provide the annotations for a balanced training (1575 clips) and validation (525 clips) split, as well as for a disjoint test set containing 525 installments from TV programs not included in the development set.</p> <p>As YouTube continuously updates, some videos may not be available in the future. Although we intend to keep ITTV updated as best as possible, please note that some content may not be available at any given time.</p> <p>Some YouTube videos (especially from the <code>Football</code> class and, to a lesser extent, the <code>Cartoons</code> class) may only be available in some countries due to regional restrictions imposed by the content creator. All videos are known to be accessible from Italy (last accessed on Nov. 25th, 2022.)<br> <br> Please contact Alessandro Ilic Mezza for further questions (e-mail: alessandroilic.mezza@polimi.it).</p>
Dataset for Automatic Refactoring Candidate Identification Leveraging Effective Code Representation
<p>The dataset consists of positive and negative case methods for Extract Method refactoring for selected GitHub repositories.<br> <br> <em>Each sample format - </em></p> <pre><code class="language-json">{ "repo_name": "...", "repo_url": "...", "positive_case_methods": ["...", "...", ...], "negative_case_methods": ["...", "...", ...] }</code></pre> <p> </p>
Semi-automatic and manual shallow landslide inventories of two extreme rainfall events.
<p>This dataset contains the polygons of automatic ( PL) and manually (ML) based shallow landslides related to two extreme rainfall events. In KML format, the dataset can be visualized on GIS software or Google Earth.</p><p>With more details, it is possible to find:</p><ul><li>AOI_2016: The study area of the extreme rainfall of November 2016, Tanerello and Arroscia Valleys NW Italy.</li><li>The 2016_PL: The inventory of potential shallow landslides semi-automatically mapped on the base of Sentinel-2 images related to extreme rainfall events that hit NW Italy in November 2016</li><li>The 2016_ML: The inventory of shallow landslides manually mapped on high-resolution images of Google Earth, related to extreme rainfall events that hit NW Italy in November 2016</li><li>AOI_2019_large: The study area of the extreme rainfall of October 2019 NW Italy.</li><li>AOI_2019: The testing area of the extreme rainfall of October 2019, Gavi Area NW Italy.</li><li>The 2019_PL_all: The inventory of potential shallow landslides semi-automatically mapped on the base of Sentinel-2 images related to extreme rainfall events that hit NW Italy in October 2019 (whole Study area)</li><li>The 2019_PL: The inventory of potential shallow landslides semi-automatically mapped on the base of Sentinel-2 images related to extreme rainfall events that hit NW Italy in October 2019 (Gavi test area)</li><li>The 2019_ML: The inventory of shallow landslides manually mapped on high-resolution images of Google Earth, related to extreme rainfall events that hit NW Italy in October 2019</li></ul><p>GEE_Script: A list of codes used in Google Earth Engine to produce NDVI time series or averaged NDVI on some sample studied areas are reported in the attached PDF. The code may be pasted and copied to the Google Earth Engine console. </p><p>The codes (if an account on Google Earth Engine is active) may be reached directly from the following URLs: </p><p><strong>Script 1. </strong>NDVI time series of some sampled areas to select the best pair of images for the PL creation (Tanarello and Arroscia Valley and GAVI AOIs; Fig. 16 of the paper). Link to GEE: <a href="https://code.earthengine.google.com/998af951fcb74519589bf8e722bb30b0?noload=true">https://code.earthengine.google.com/998af951fcb74519589bf8e722bb30b0?noload=true</a></p><p><strong>Script 2.</strong> sampled NDVI time series from different intersection cases for the Tanarello and Arroscia Valley study area (2016 Event). Link to GEE: <a href="https://code.earthengine.google.com/b622cb64f90771ced78ef73bad9cc50f?noload=true">https://code.earthengine.google.com/b622cb64f90771ced78ef73bad9cc50f?noload=true</a></p><p><strong>Script 3. </strong>Sampled NDVI time series from different land-use cases for the Gavi study area (2019 Event). Link to GEE: <a href="https://code.earthengine.google.com/f686c60b78a3dee0b2a2c94a259ccff2?noload=true">https://code.earthengine.google.com/f686c60b78a3dee0b2a2c94a259ccff2?noload=true</a></p><p><strong>Script 4.</strong> Multi-temporal-averaged NDVIvar Link to GEE Script: <a href="https://code.earthengine.google.com/bfc2e570bb675372c4c482eef682be4a?noload=true">https://code.earthengine.google.com/bfc2e570bb675372c4c482eef682be4a?noload=true</a> for the whole Gavi study area (2019 flood) and <a href="https://code.earthengine.google.com/89e1c0a1361860cd407b7e6ab8bb95de?noload=true">https://code.earthengine.google.com/a3390b262cef1b5f42837c88d8791b5b?noload=true</a> for the entire Arroscia-Tanarello study area</p><p>The full description of the methodology can be found in the paper of Notti et al., 2023</p><p>Notti, D., Cignetti, M., Godone, D., and Giordan, D.: Semi-automatic mapping of shallow landslides using free Sentinel-2 images and Google Earth Engine, Nat. Hazards Earth Syst. Sci., 23, 2625–2648, <a href="https://doi.org/10.5194/nhess-23-2625-2023">https://doi.org/10.5194/nhess-23-2625-2023</a>, 2023</p>
Input and output data (images + boulder labels, model setup, model weights and more) for the manuscript "Automatic characterization of boulders on planetary surfaces from high-resolution satellite images"
<p><strong>File 1:</strong> raw_data_BOULDERING.zip</p> <p><strong>Size:</strong> 8.8 GB</p> <p><strong>Summary: </strong>It contains all of the rasters (planetary images) and labeled boulders (raw data):</p> <ul> <li> <p>a boulder-mapping file, which is the manually digitized outline of boulders.</p> </li> <li> <p>a ROM file (stands for Region of Mapping), which depicts the image patches on which the boulder mapping has been conducted.</p> </li> <li> <p>a global-tiles file, which shows all of the image patches within a raster.</p> </li> </ul> <p>There are multiple locations/images per planetary body.</p> <p><strong>Structure:</strong></p> <pre>. └── raw_data/ ├── earth/ │ └── image_name/ │ ├── shp/ │ │ ├── <image_name>-ROM.shp │ │ ├── <image_name>-boulder-mapping.shp │ │ └── <image_name>-global-tiles.shp │ └── raster/ │ └── <image_name>.tif ├── mars/ │ └── image_name/ │ ├── shp/ │ │ ├── <image_name>-ROM.shp │ │ ├── <image_name>-boulder-mapping.shp │ │ └── <image_name>-global-tiles.shp │ └── raster/ │ └── <image_name>.tif └── moon/ └── image_name/ ├── shp/ │ ├── <image_name>-ROM.shp │ ├── <image_name>-boulder-mapping.shp │ └── <image_name>-global-tiles.shp └── raster/ └── <image_name>.tif</pre> <p> </p> <p><strong>File 2:</strong> best_model.zip</p> <p><strong>Size:</strong> 624.7 MB</p> <p><strong>Summary:</strong></p> <p>This zip file contains all of the inputs and outputs required/obtained from the training of the BoulderNet Mask R-CNN model (model setup, augmentation pipeline, model weights, log during training, logged metrics):</p> <ul> <li> <p>augmentation_pipeline.json (required as inputs for the training of the algorithm to apply augmentations). See <a href="https://github.com/astroNils">https://github.com/astroNils</a> and the MLtools repository for more information.</p> </li> </ul> <ul> <li> <p>Base-RCNN-FPN.yaml (base model setup file).</p> </li> <li> <p>config.yaml (complete model setup file, merge of the base and Mars-Moon-Earth setup file).</p> </li> <li> <p>Mars-MoonEarth-v050...yaml (model setup file).</p> </li> <li> <p>log.txt (log during training of the algorithm).</p> </li> <li> <p>model_0055999.pth (model weights at second last saving step)</p> </li> <li> <p>model_0063999.pth (model weights at last saving step)</p> </li> </ul> <p>We advice the use of model weights model_0055999.pth (to avoid slight overfitting).</p> <p><strong>File 3:</strong> Apr2023-Mars-Moon-Earth-mask-5px.zip (pre-processed input images)</p> <p><strong>Size:</strong> 252.8 MB</p> <p><strong>Summary:</strong></p> <p>This zip files contains the input data (images and boulder outlines) for the train, validation and test datasets. See <a href="https://github.com/astroNils">https://github.com/astroNils</a> and the MLtools repository for more information in how-to-use the different files.</p> <ul> <li> <p>The json folder contains json files that can be given as input (as a custom dataset) to the Detectron2 platform. The only differences between the two files is how the bounding boxes around masks have been generated. We advised to use "Apr2023-Mars-Moon-Earth-mask-5px.json".</p> </li> <li> <p>The pkl folder and pickle file includes some informations about the 950 image patches in our boulder dataset.</p> </li> <li> <p>The pre-processing folder contains all of the training, validation and test image patches and corresponding shapefiles.</p> </li> <li> <p>The shapefile folder is actually empty (it should not be there!).</p> </li> </ul> <p><strong>Structure:</strong></p> <pre>. └── preprocessed_inputs/ ├── json ├── pkl ├── preprocessing/ │ ├── train/ │ │ ├── images │ │ └── labels │ ├── validation/ │ │ ├── images │ │ └── labels │ └── test/ │ ├── images │ └── labels └── shp</pre> <p> </p>
ScienceDex guides
Understand access before you commit
These curated guides explain access requirements, typical timelines, costs, and reuse considerations for widely used research datasets.
Allen Brain Atlas
Allen Brain Atlas is an Allen Institute collection of brain map atlases, datasets, APIs, and analysis tools covering mouse, human, and non-human primate brain resources.
Annotated Behaviour and Observability Dataset (ABODe)
ABODe is a University of Edinburgh DataShare dataset for behavior classification in group-housed mice using home-cage video, identities, bounding boxes, ground-plate positions, and annotator labels.
DANDI Archive for NWB datasets
DANDI is a BRAIN Initiative archive for publishing and sharing neurophysiology data, including electrophysiology, optophysiology, and behavioral data packaged as NWB and related standards.
International Brain Laboratory public data
The International Brain Laboratory public data releases expose standardized mouse decision-making experiments, including Neuropixels recordings, widefield calcium imaging, behavior, and session metadata accessed through the ONE API.
OpenNeuro
OpenNeuro is a free, open platform for sharing neuroimaging datasets, with public search, dataset pages, and download paths for web, S3, DataLad, and the OpenNeuro CLI.