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.
46
datasets available to search
ShareScore release 0.9.0
Dataset results
46 results for “unsupervised learning”
Detecting repeating earthquakes on the San Andreas Fault with unsupervised machine-learning of spectrograms (supplementary material)
<p>Supplementary material for Sawi et al., 2023, <i>Detecting repeating earthquakes on the San Andreas Fault with unsupervised machine-learning of spectrograms </i>(The Seismic Record). Catalog of repeating earthquakes in sequences on a 10-km long segment of the San Andreas Fault in California from 1984-2019. </p><p> </p><p><strong>Catalog Header</strong></p><p>YR/MO/DY...........Date of event</p><p>HR/MN/SC...........Time of event</p><p>LAT/LON/DEP........Location of event</p><p>EX/EY/EZ...........Relative location uncertainty (in m)</p><p>MAG................NCSN magnitude</p><p>evID.................NCSN event ID</p><p>seqID................Repeating earthquake sequence ID</p><p>isRESp............Is quasi-periodic RES (bool)</p><p> </p><p><strong>References: </strong></p><p>Sawi T., Waldhauser F., Holtzman B. K., Groebner, N. (2023) Detecting repeating earthquakes on the San Andreas Fault with unsupervised machine-learning of spectrograms. The Seismic Record. </p><p>Waldhauser, F., and Schaff, D. P. (2021). A Comprehensive Search for Repeating Earthquakes in Northern California: Implications for Fault Creep, Slip Rates, Slip Partitioning, and Transient Stress. J Geophys Res B Solid Earth, 126(11), 1–22. <a href="https://doi.org/10.1029/2021JB022495">https://doi.org/10.1029/2021JB022495</a></p>
Datasets for "Mapping Lunar Swirls with Machine Learning: The Application of Unsupervised and Supervised Classification Algorithms in Reiner Gamma and Mare Ingenii"
<p>Final surface reflectance data at 2.6 m/pixel resolution with floating point values are available as GeoTiff and ASCII text files. Definition files for the K-Means and MLC algorithms in classifying swirl units are also available as ASCII text files. See README file for further details.</p> <p>Data used in the research article:</p> <p>Chuang, F.C., M.D. Richardson, J.R. Weirich, A.A. Sickafoose, and D.L. Domingue, 2022. Mapping Lunar Swirls with Machine Learning: The Application of Unsupervised and Supervised Image Classification Algorithms in Reiner Gamma and Mare Ingenii. The Planetary Science Journal, 3:231. doi://10.3847/PSJ/ac8f43</p> <p> </p>
Data bundle for "Spherical-angular dark field imaging and sensitive microstructural phase clustering with unsupervised machine learning"
<p>Prepared by Tom McAuliffe (t.mcauliffe17@imperial.ac.uk)</p> <p>This repository is a release of the raw data and analysis results for: 'Spherical-angular dark field imaging and sensitive microstructural phase clustering with unsupervised machine learning' </p> <p>The raw data is given as 'yprime.h5' - this contains patterns and metadata in the Bruker-exported format.</p> <p>Scripts for dataset decomposition into latent factors are given in 'Scripts'.</p> <p>Our spherical analysis code is included in 'SphericalAngleDF'.</p> <p>Outputs of our analysis code are contained in 'Analysis'.</p> <p>Figures for the paper are included in 'Figures'.</p> <p> </p>
Automated MESSENGER Plasma Region Classifications via Unsupervised Transfer Learning
<p>This file contains the 1-minute resolution dataset (“labeled_sunside_data_3labels.csv”) for Toy-Edens et al.’s Automated Classification of MESSENGER Plasma Observations via Unsupervised Transfer Learning. The 1-minute resolution file contains the rolled up 1-minute epoch, features that go into clustering and post-cleaning methods, spacecraft positions (in MSO), total magnetic field, raw and cleaned clustering labels, and raw and cleaned transition name.</p> <p>We ask that if you use any parts of the dataset that you cite Toy-Edens et al.’s Automated Classification of MESSENGER Plasma Observations via Unsupervised Transfer Learning (DOI: 10.3389/fspas.2025.1608091).</p> <p>This work was supported by NASA grants 80NSSC19K0789 and 80NSSC22K0993.</p> <p> </p> <p>The following tables detail the contents of the described files:</p> <p><strong>labeled_sunside_data_3labels.csv description</strong></p> <table style="width: 100.063%; height: 851.2px;"> <tbody> <tr style="height: 47.6px;"> <td style="width: 17.3792%; height: 47.6px;"> <p><strong>Column Name</strong></p> </td> <td style="width: 78.9512%; height: 47.6px;"> <p><strong>Description</strong></p> </td> </tr> <tr style="height: 47.6px;"> <td style="width: 17.3792%; height: 47.6px;"> <p> Epoch</p> </td> <td style="width: 78.9512%; height: 47.6px;"> <p>Epoch in datetime (YYYY-MM-DD HH:MM:SS)</p> </td> </tr> <tr style="height: 47.6px;"> <td style="width: 17.3792%; height: 47.6px;"> <p> x_mso</p> </td> <td style="width: 78.9512%; height: 47.6px;"> <p>x position of the spacecraft in MSO [km]</p> </td> </tr> <tr style="height: 47.6px;"> <td style="width: 17.3792%; height: 47.6px;"> <p> y_mso</p> </td> <td style="width: 78.9512%; height: 47.6px;"> <p>y position of the spacecraft in MSO [km]</p> </td> </tr> <tr style="height: 47.6px;"> <td style="width: 17.3792%; height: 47.6px;"> <p> z_mso</p> </td> <td style="width: 78.9512%; height: 47.6px;"> <p>z position of the spacecraft in MSO [km]</p> </td> </tr> <tr style="height: 47.6px;"> <td style="width: 17.3792%; height: 47.6px;"> <p> btot_mso</p> </td> <td style="width: 78.9512%; height: 47.6px;"> <p>Total magnetic field [nT]</p> </td> </tr> <tr style="height: 47.6px;"> <td style="width: 17.3792%; height: 47.6px;"> <p> norm_Btot</p> </td> <td style="width: 78.9512%; height: 47.6px;"> <p>Magnitude of the total magnetic field normalized to 150nT. See paper for more information</p> </td> </tr> <tr style="height: 67.2px;"> <td style="width: 17.3792%; height: 67.2px;"> <p> ratio_max_width</p> </td> <td style="width: 78.9512%; height: 67.2px;"> <p>Ratio of the width of the most prominent ion spectra peak (in number of energy channels) to max number of energy channels. See paper for more information</p> </td> </tr> <tr style="height: 67.2px;"> <td style="width: 17.3792%; height: 67.2px;"> <p> ratio_high_low</p> </td> <td style="width: 78.9512%; height: 67.2px;"> <p>Ratio of the mean of the log intensity of high energies in the ion spectra to the mean of the log intensity of low energies in the ion spectra. See paper for more information</p> </td> </tr> <tr style="height: 67.2px;"> <td style="width: 17.3792%; height: 67.2px;"> <p> high_intensity</p> </td> <td style="width: 78.9512%; height: 67.2px;"> <p>Boolean if there is a peak with a higher minimum intensity threshold. See paper for more information</p> </td> </tr> <tr style="height: 67.2px;"> <td style="width: 17.3792%; height: 67.2px;"> <p> spectra_counts</p> </td> <td style="width: 78.9512%; height: 67.2px;"> <p>A ratio of spectra bins with non-zero counts to all possible spectra bins (i.e. way to determine if too much missing spectra data). See paper for more information</p> </td> </tr> <tr style="height: 67.2px;"> <td style="width: 17.3792%; height: 67.2px;"> <p> raw_named_label</p> </td> <td style="width: 78.9512%; height: 67.2px;"> <p>Raw cluster assigned plasma region label (allowed values: magnetosheath, magnetosphere, solar wind)</p> </td> </tr> <tr> <td style="width: 17.3792%;"> <p>intermediate_named_label</p> </td> <td style="width: 78.9512%;"> <p>Cleaned cluster assigned plasma region label with only relabeling rules applied. See paper for more information</p> </td> </tr> <tr style="height: 67.2px;"> <td style="width: 17.3792%; height: 67.2px;"> <p> named_label</p> </td> <td style="width: 78.9512%; height: 67.2px;"> <p>Cleaned cluster assigned plasma region label with relabeling rules and post-processing applied (use these unless have a specific reason to use raw labels). See paper for more information</p> </td> </tr> <tr style="height: 47.6px;"> <td style="width: 17.3792%; height: 47.6px;"> <p> raw_transition_name</p> </td> <td style="width: 78.9512%; height: 47.6px;"> <p>Raw transition names (e.g. bow shock, magnetopause) based on "raw_named_label" cluster labels. See paper for more information</p> </td> </tr> <tr style="height: 67.2px;"> <td style="width: 17.3792%; height: 67.2px;"> <p> transition_name</p> </td> <td style="width: 78.9512%; height: 67.2px;"> <p>Cleaned transition names (e.g. bow shock, magnetopause) after removing likely transient transitions based on "named_label" cluster labels. See paper for more information</p> </td> </tr> </tbody> </table> <p> </p>
Supporting DNN Safety Analysis and Retraining through Heatmap-based Unsupervised Learning
<p>This repository provides the data used for the experiments of the paper "Supporting DNN Safety Analysis and Retraining through Heatmap-based Unsupervised Learning" by Hazem Fahmy, Fabrizio Pastore, Mojtaba Bagherzadeh, and Lionel Briand appearing in IEEE Transactions on Reliability (doi: 10.1109/TR.2021.3074750)</p> <p>Deep neural networks (DNNs) are increasingly important in safety-critical systems, for example in their perception layer to analyze images. Unfortunately, there is a lack of methods to ensure the functional safety of DNN-based components.</p> <p>We observe three major challenges with existing practices regarding DNNs in safety-critical systems: (1) scenarios that are underrepresented in the test set may lead to serious safety violation risks, but may, however, remain unnoticed; (2) char- acterizing such high-risk scenarios is critical for safety analysis; (3) retraining DNNs to address these risks is poorly supported when causes of violations are difficult to determine.</p> <p>To address these problems in the context of DNNs analyzing images, we propose HUDD, an approach that automatically supports the identification of root causes for DNN errors. HUDD identifies root causes by applying a clustering algorithm to heatmaps capturing the relevance of every DNN neuron on the DNN outcome. Also, HUDD retrains DNNs with images that are automatically selected based on their relatedness to the identified image clusters.</p> <p>We evaluated HUDD with DNNs from the automotive domain. HUDD was able to identify all the distinct root causes of DNN errors, thus supporting safety analysis. Also, our retraining approach has shown to be more effective at improving DNN accuracy than existing approaches.</p> <p> </p>
UNSUPERVISED MACHINE LEARNING AND VECTOR MODELS IN DESIGNING AND OPTIMIZATION OF TELECOM RETAIL CHANNELS
<p>This paper examines the use of unsupervised machine learning and vector models in the design and optimization of retail channels for telecommunications services. Unsupervised machine learning allows you to analyze and identify hidden patterns in large volumes of untagged data, which is especially important in a dynamically changing consumer market. Vector models, in turn, provide high accuracy of demand forecasting and inventory management, contributing to an increase in the efficiency of trading channels. The synergy of these technologies allows companies to improve customer experience, optimize operational processes and increase competitiveness in the market. The main focus of the work is on data processing methods, including correlation analysis, the use of the support vector machine (SVM) method and its adaptation to solve problems related to predicting customer behavior and optimizing logistics processes.</p>
Learning Unsupervised Hierarchies of Audio Concepts
<p>Dataset of our paper <a href="https://arxiv.org/abs/2207.11231">"Learning Unsupervised Hierarchies of Audio Concepts"</a> published at the <a href="https://ismir2022.ismir.net/">ISMIR 2022</a> conference.</p> <p>For usage examples, please refer to our code repository at <a href="https://github.com/deezer/concept_hierarchy">github.com/deezer/concept_hierarchy</a>.</p> <p><strong>Paper abstract</strong></p> <p>Music signals are difficult to interpret from their low-level features, perhaps even more than images: e.g. highlighting part of a spectrogram or an image is often insufficient to convey high-level ideas that are genuinely relevant to humans. In computer vision, concept learning was therein proposed to adjust explanations to the right abstraction level (e.g. detect clinical concepts from radiographs). These methods have yet to be used for MIR.<br> In this paper, we adapt concept learning to the realm of music, with its particularities. For instance, music concepts are typically non-independent and of mixed nature (e.g. genre, instruments, mood), unlike previous work that assumed disentangled concepts. We propose a method to learn numerous music concepts from audio and then automatically hierarchise them to expose their mutual relationships. We conduct experiments on datasets of playlists from a music streaming service, serving as a few annotated examples for diverse concepts. Evaluations show that the mined hierarchies are aligned with both ground-truth hierarchies of concepts -- when available -- and with proxy sources of concept similarity in the general case.</p> <p><strong>Citation</strong></p> <p>If you use this material, please consider citing our work with the following:</p> <pre><code>@inproceedings{afchar2022learning, title={Learning Unsupervised Hierarchies of Audio Concepts}, author={Afchar, Darius and Hennequin, Romain and Guigue, Vincent}, booktitle={International Society of Music Information Retrieval Conference (ISMIR)}, year={2022} }</code></pre> <p><strong>How can I send you a love letter?</strong></p> <p>You can contact us at <a href="mailto:research@deezer.com?subject=Dataset%20ISMIR%20Afchar">research@deezer.com</a>.</p>
Data and scripts from "Unsupervised learning for structure detection in plastically deformed crystals"
<p>This documents contains the scripts and dataset used for the paper "Unsupervised learning for structure detection in plastically deformed crystals".</p> <p> </p> <p>More precisely it contains 4 folders :</p> <p><br> DumpForFigures : subfolder containing the atomic positions in .dump format (see lammps documentation) used for the article figures.</p> <p>DumpForTraining : subfolder containing the atomic position in .dump format (see lammps documentation) used for training the autoencoder.</p> <p>ScriptsToDetectStructuresFromDump : subfolder containing the script sused to detect the substructures of the system by combining autoencoder and clustering methods. This folder contains a readme with the details of the contents.</p> <p>ScriptToGenerateDump : subfolder containing the scripts used to generate the atomic data with molecular dynamics. These data are then used to train the autoencoder. This folder contains a readme with the details of the contents.</p> <p>REQUIREMENTS :</p> <p> </p> <p>Lammps</p> <p>Python3 with packages :</p> <p>-numpy</p> <p>-matplotlib</p> <p>-pyscal</p> <p>-sci-kit learn</p> <p>-pytorch</p> <p>-glob</p> <p> </p> <p> </p> <p> </p>
Illuminating the Hierarchical Segmentation of Faults through an Unsupervised Learning Approach applied to clouds of earthquake hypocenters [Dataset]
<p>Data repository to the preprint “Illuminating the Hierarchical Segmentation of Faults through an Unsupersived Larning Approach applied to clouds of earthquake hypocenters” by Piegari et al. (2023), including the datasets for the three analyzed earthquake catalogs.</p>
Learning Unsupervised Knowledge-Enhanced Representations to Reduce the Semantic Gap in Information Retrieval (Evaluation datasets)
<p>This dataset contains all the runs, pools, plots and analyses to reproduce the results presented in the paper: "Learning Unsupervised Knowledge-Enhanced Representations to Reduce the Semantic Gap in Information Retrieval ", 2020.</p>
Molecular dynamics simulations data for "Bayesian unsupervised learning reveals hidden structure in concentrated electrolytes".
<p>Molecular dynamics simulation data created and used in "Bayesian unsupervised learning reveals hidden structure in concentrated electrolytes".</p> <p> </p>
Data from: A demonstration of unsupervised machine learning in species delimitation
One major challenge to delimiting species with genetic data is successfully differentiating population structure from species-level divergence, an issue exacerbated in taxa inhabiting naturally fragmented habitats. Many fields of science are now using machine learning, and in evolutionary biology supervised machine learning has recently been used to infer species boundaries. These supervised methods require training data with associated labels. Conversely, unsupervised machine learning (UML) uses inherent data structure and does not require user-specified training labels, potentially providing more objectivity in species delimitation. Here we demonstrate the utility of three UML approaches (random forests, variational autoencoders, t-distributed stochastic neighbor embedding) for species delimitation in an arachnid taxon with high population genetic structure (Opiliones, Laniatores, Metanonychus). We find that UML approaches successfully cluster samples according to species-level divergences and not high levels of population structure, while model-based validation methods severely over-split putative species. UML offers intuitive data visualization in two-dimensional space, the ability to accommodate various data types, and has potential in many areas of systematic and evolutionary biology. We argue that machine learning methods are ideally suited for species delimitation and may perform well in many natural systems and across taxa with diverse biological characteristics.
Data from: Unsupervised machine learning reveals mimicry complexes in bumble bees occur along a perceptual continuum
Müllerian mimicry theory states that frequency dependent selection should favour geographic convergence of harmful species onto a shared colour pattern. As such, mimetic patterns are commonly circumscribed into discrete mimicry complexes each containing a predominant phenotype. Outside a few examples in butterflies, the location of transition zones between mimicry complexes and the factors driving mimicry zones has rarely been examined. To infer the patterns and processes of Müllerian mimicry, we integrate large-scale data on the geographic distribution of colour patterns of social bumble bees across the contiguous United States and use these to quantify colour pattern mimicry using an innovative, unsupervised machine learning approach based on computer vision. Our data suggest that bumble bees exhibit geographically clustered, but sometimes imperfect colour patterns and that mimicry patterns gradually transition spatially, rather than exhibit discrete boundaries. Additionally, examination of colour pattern transition zones of three comimicking, polymorphic species, where active selection is driving phenotype frequencies, revealed their transition zones to differ in location within a broad region of poor mimicry. Potential factors influencing mimicry transition zone dynamics are discussed.
Pairwise Learning using Unsupervised Bottleneck Features for Zero-Resource Speech Challenge 2017 (System 1)
<p>The system is for track1 alone. We trained an antoencoder using unsupervised bottleneck features with word-pair information from Switchboard. The unsupervised bottleneck features was extracted from an extractor of multi-task learning deep neural networks (MTL-DNN). The word-pair information was the ground truth from Switchboard. The final features are obtained from the third layer in our pairwise trained autoencoder.</p>
Pairwise Learning using Unsupervised Bottleneck Features for Zero-Resource Speech Challenge 2017 (System 3)
<p>The system is for track1 alone. We trained an antoencoder using unsupervised bottleneck features with word-pair information from unsupervised term detection (UTD) on all corpora of five languages. The unsupervised bottleneck features was extracted from an extractor of multi-task learning deep neural networks (MTL-DNN). The word-pair was found by UTD. The UTD process was built on ZRTools. The final features are obtained from the third layer in our pairwise trained autoencoder.</p>
Pairwise Learning using Unsupervised Bottleneck Features for Zero-Resource Speech Challenge 2017 (System 2)
<p>The system is for track1 alone. We trained an antoencoder using unsupervised bottleneck features with word-pair information from unsupervised term detection (UTD) only on the give ENGLISH corpus. The unsupervised bottleneck features was extracted from an extractor of multi-task learning deep neural networks (MTL-DNN). The word-pair was found by UTD. The UTD process was built on ZRTools. The final features are obtained from the third layer in our pairwise trained autoencoder.</p>
Research data supporting: "Classifying soft self-assembled materials via unsupervised machine learning of defects"
<p>Research data supporting: "Classifying soft self-assembled materials via unsupervised machine learning of defects".</p> <p>The root folder contains 5 folders:</p> <ol> <li>FIBERS</li> <li>MEMBRANES_and_MICELLES</li> <li>NANOPARTICLES</li> <li>COMPARISON</li> <li>paper_images</li> </ol> <p>The folders 1. to 3. contain the data for every soft-matters architecture used to produce the results discussed in the main paper. Each of these folders contain additional sub-fordels: TRAJ, SOAP, PCA, CLUSTERING, containing the files discussed in the main paper.</p> <p>Folder 4. contains the data of the comparison between different classes of materials (SOAP, PCA, and CLUSTERING sub-folders).</p> <p>Folder 5. contains the images that are showed in the main paper and in the Supporting Information.</p>
Comparing Storm Resolving Models and Climates via Unsupervised Machine Learning
<p>Storm-resolving climate models (SRMs) have gained international interest for their unprecedented detail with which they globally resolve convection. However, this high resolution also makes it difficult to quantify the emergent differences or similarities among complex atmospheric formations induced by different parameterizations of sub-grid information. This paper uses modern unsupervised machine learning methods to analyze and intercompare SRMs based on their high-dimensional simulation data, learning low-dimensional latent representations and assessing model similarities based on these representations. To quantify such inter-SRM ``distribution shifts'', we use variational autoencoders in conjunction with vector quantization. Our analysis involving nine different global SRMs reveals that only six of them are aligned in their representation of atmospheric dynamics. Our analysis furthermore reveals regional and planetary signatures of the convective response to global warming in a fully unsupervised, data-driven way. In particular, this approach can help elucidate the effects of climate change on rare convection types, such as ``Green Cumuli''. Our study provides a path toward evaluating future high-resolution global climate simulation data more objectively and with less human intervention than has historically been needed.</p>
Observing flow of He II with unsupervised machine learning
<p>Data repository for observing flow in He II with unsupervised machine learning.</p>
Preprocessed Data for "Comparing Storm Resolving Models and Climates via Unsupervised Machine Learning"
<p>Preprocessed Data (training and test) for 3 SRMs used in "Comparing Storm Resolving Models and Climates via Unsupervised Machine Learning". Here we included ICON, SPCAM, and SPCAM with sea surface tem[eratures warmed by +4K. Additionally we include lat/lon information for the test data.</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.