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.
2,025
datasets available to search
ShareScore release 0.9.0
Dataset results
2,025 results for “AIS”
2003-2005 Dataset [2/7] for the models trained and tested in the paper 'Can AI be enabled to dynamical downscaling? Training a Latent Diffusion Model to mimic km-scale COSMO-CLM downscaling of ERA5 over Italy'
<p>This repository contains part 2/7 of the full dataset used for the models of the <a href="https://arxiv.org/abs/2406.13627">preprint</a> "Can AI be enabled to dynamical downscaling? Training a Latent Diffusion Model to mimic km-scale COSMO-CLM downscaling of ERA5 over Italy". </p> <p>This dataset comprises 3 years of normalized hourly data for both low-resolution predictors [16 km] and high-resolution target variables [2km] (2mT and 10-m U and V), from 2003-2005. Low-resolution data are preprocessed ERA5 data while high-resolution data are preprocessed VHR-REA CMCC data. Details on the performed preprocessing are available in the paper.</p> <p>To use the data, clone the corresponding <a href="https://github.com/DSIP-FBK/DiffScaler">repository</a>, unzip this zip file in the data folder, and download from Zenodo the other parts of the dataset listed in the related works.</p>
Automated rationale generation: a technique for explainable AI and its effects on human perceptions (Dataset)
<p>Explainable AI Dataset for paper published in the proceedings of IUI 2019 titled, "<strong>Automated rationale generation: a technique for explainable AI and its effects on human perceptions</strong>". Consists of data collected from human participants via Mechanical Turk.<br> <strong>Abstract</strong>:<br> <em>Automated rationale generation</em> is an approach for real-time explanation generation whereby a computational model learns to translate an autonomous agent's internal state and action data representations into natural language. Training on human explanation data can enable agents to learn to generate human-like explanations for their behavior. In this paper, using the context of an agent that plays <em>Frogger</em>, we describe (a) how to collect a corpus of explanations, (b) how to train a neural rationale generator to produce different styles of rationales, and (c) how people perceive these rationales. We conducted two user studies. The first study establishes the plausibility of each type of generated rationale and situates their user perceptions along the dimensions of <em>confidence, humanlike-ness, adequate justification, and understandability.</em> The second study further explores user preferences between the generated rationales with regard to <em>confidence</em> in the autonomous agent, communicating <em>failure and unexpected behavior.</em> Overall, we find alignment between the intended differences in features of the generated rationales and the perceived differences by users. Moreover, context permitting, participants preferred detailed rationales to form a stable mental model of the agent's behavior.</p>
ai-matrix NCF
<p>Large files in ai-matrix.</p>
ai-matrix BERT_PyTorch
<p>ai-matrix BERT_PyTorch pretrained model, converted from TensorFlow model, <strong>uncased_L-24_H-1024_A-16</strong></p>
Protein structure data for "AI-predicted protein deformation encodes energy landscape perturbation"
<p>AF2-predicted protein structures of WT and mutant proteins that have corresponding ddG measurements in the ThermoMutDB database of protein mutant stability measurements. PDB structures are compressed using <a href="https://github.com/steineggerlab/foldcomp/">FoldComp</a>, and saved in "structures.zip".</p> <p>Summary of the final dataset and results can be found in "results_summary.pkl".</p> <p>Code used to plot figures can be found in "code4figs.zip".</p>
Automated Generation of Code Contracts - Generative AI to the Rescue?
<p>This replication package provides the setup and results to generate OpenJML code contracts for Java source code by fine-tuning and employing the resulting CodeT5 and CodeT5+ transformer models. Our code contract generation setup involved the training of the AI models and application. Furthermore, we analyzed the generated annotations wrt. thier logical validity and the type of OpenJML compilation errors. Both methods, together with the results are similarly provided.</p> <p><strong>Source Code Repository (see also scripts-sources.tar): </strong></p> <ul> <li><a href="https://github.com/SEG-UNIBE/auto-generated-code-contracts">https://github.com/SEG-UNIBE/auto-generated-code-contracts</a></li> <li><a href="../doi/10.5281/zenodo.13356451">https://zenodo.org/doi/10.5281/zenodo.13356451</a></li> </ul> <p><strong>Replication Package: </strong>contains the following [folders]</p> <ul> <li><strong>Scripts:</strong> <ul> <li>[scripts-sources.tar]: source codes of the following scripts <ul> <li>Python scripts that we used for training and adding the OpenJML code contracts to the Java methods</li> <li>automated analyses of the studied source code classes and the type of compilation errors</li> </ul> </li> </ul> </li> <li><strong>Sourcegraph Search Results:</strong> <ul> <li>[sourcegraph-results.tar]: the results of the Sourcegraph search queries </li> </ul> </li> <li><strong>Datasets:</strong> <ul> <li>[dataset.tar]: the dataset including the weka-project which contributes two-thirds of the contracts</li> <li>[dataset-withoutweka.tar]: the dataset without weka, which is significantly smaller and was used to examine the performance bias when training and testing without weka</li> </ul> </li> <li><strong>CodeT5 Models:</strong><br> <ul> <li>[codet5-contracts.tar]: the best performing CodeT5 model which was fine-tuned to create OpenJML annotations for methods</li> <li>[codet5p-contracts.tar]: the best performing CodeT5+ model which was fine-tuned to create OpenJML annotations for methods</li> <li>[codet5p-contracts-withoutweka.tar]: the CodeT5+ model which was trained without weka on the same task</li> </ul> </li> <li><strong>Analysis Results:</strong> <ul> <li>[analysis-results.tar/compilability-analysis]: the results of the compilability analysis <ul> <li>the subjects to which we applied the best performing CodeT5+</li> <li>the compilation results and their analysis</li> </ul> </li> <li>[analysis-results.tar/logical-analysis] the results of the logical analysis <ul> <li>the analysis of logic validity of SimpleStack and SimpleTicTacToe</li> </ul> </li> </ul> </li> </ul>
Presentation AI and Facial Recognition V2
Open the record for dataset details and reuse information.
AI in Climate Change
<p><span>Climate change is an urgent issue that must be addressed by individuals, communities, governments, and organizations around the world.<span> </span>Its immediate effects include extreme weather events and unpredictable rainfall, impacting various sectors. Aligned with UN Sustainable Development Goals, addressing climate change is paramount. Despite deteriorating environmental conditions, technology continues to improve. It might be able to withstand and potentially even reduce the effects of climate change on the environment. Artificial intelligence is a technology that has been evolving quickly during the past years. Artificial intelligence is present in many aspects of our daily lives, such as search recommendations on social media. This systematic literature review examines 54 referenced papers, utilizing the Kitchenham approach to validate five research questions. According to the statistics, the Random Forest technique was employed in 18 out of 54 studies to build artificial intelligence for climate change issues. China emerged as the leader in conducting studies on AI's role in addressing climate change challenges. Within climate change research, hydrology stands out as a prominent and extensively discussed aspect. Overall, AI for Climate Change has made considerable improvements, highlighting the significance of continuing study in this specific area. </span></p>
AI Platforms Overview for Higher Education
<p>This release includes a detailed table comparing various AI platforms. The table covers:</p> <ul> <li>Focus: Primary applications of each platform.</li> <li>Strengths: Key features and benefits.</li> <li>Uniqueness: Distinctive aspects.</li> <li>Limitations: Notable constraints.</li> </ul> <p><strong>How to Use</strong></p> <ol> <li>Download the attached files.</li> <li>Review the table for a comprehensive comparison.</li> </ol> <p><strong>DOI</strong></p> <p>For more information, visit the DOI page: [10.5281/zenodo.13621693](https://doi.org/10.5281/zenodo.13621693)</p> <p><strong>Contact</strong></p> <p>For any inquiries, contact <a href="mailto:bsrinivasan@twu.edu">bsrinivasan@twu.edu</a>.</p>
The Impact of Generative AI-Powered Code Generation Tools on Software Engineer Hiring: Recruiters' Experiences, Perceptions, and Strategies
<p>This is the dataset for the paper: The Impact of Generative AI-Powered Code Generation Tools on Software Engineer Hiring: Recruiters' Experiences, Perceptions, and Strategies</p> <p> This paper was accepted for publication at the 58th Hawaii International Conference on System Sciences (HICSS) - Software Technology Track</p>
FIGURE 3. Phyllagathis rajah A in Two new taxa of Melastomataceae Trib. Sonerileae: Phyllagathis rajah and Sonerila metallica from Batang Ai, Sarawak, Borneo
FIGURE 3. Phyllagathis rajah A. Habit and habitat; B. Leaf abaxial surface; C. Leaf adaxial surface; D & E. Inflorescence; F. Flower, top view; G. Flower, side view; H. Cross section of ovary, showing crown; I. & J. Equatorial view of pollen grain (SEM); K. Polar view of
FIGURE 5. Sonerila metallica A in Two new taxa of Melastomataceae Trib. Sonerileae: Phyllagathis rajah and Sonerila metallica from Batang Ai, Sarawak, Borneo
FIGURE 5. Sonerila metallica A. Habit and habitat; B. Cultivated plant at anthesis; B. S. nervulosa at anthesis;C. Leaf abaxial surface; D. Leaf adaxial surface; E. Flower, top view; F. Flower, side view; G. Fruit, side view; H. Fruit, top view; I. Seed; J. Seed, showing papillae and tubercles on dorsal surface; K. Pollen grain, equatorial view (SEM); L. Pollen grain, polar view (SEM); M. Pollen grain, polar view
FIGURE 4. Sonerila metallica A in Two new taxa of Melastomataceae Trib. Sonerileae: Phyllagathis rajah and Sonerila metallica from Batang Ai, Sarawak, Borneo
FIGURE 4. Sonerila metallica A. Habit; B,B'. Portion of larger leaf adaxial and abaxial surface, showing distance from lateral veins to margin; C,C'. Smaller leaf adaxial surface, showing leaf variation, C''. Smaller leaf, abaxial surface; D–D'''. Bracts, showing variation; E, E'. Flower, top and side view; F. Stamens and filaments Ventral view, F'. drosal.view, F''. Side view; G. Style; H, H'. Petal, showing adaxial and abaxial surface; I. Cross section of ovary; J. Immature fruit, J'. Mature fruit, J''. Fruit, top view.
FIGURE 2. Phyllagathis rajah A in Two new taxa of Melastomataceae Trib. Sonerileae: Phyllagathis rajah and Sonerila metallica from Batang Ai, Sarawak, Borneo
FIGURE 2. Phyllagathis rajah A. Habit; B,B'. Portion of leaf adaxial and abaxial surfaces, also showing leaf margin; C. Inflorescence; D. Bract; E,E'. Flower, front view and side view; F. Petal; G. Stamen and filament, ventral view, G'. Dorsal view, G''. Side view; H. Style; I. Hypanthium and calyx; J. Cross section of ovary (all from the type, C.-W. Lin 550, TAIF).
FIGURE 1 in Two new taxa of Melastomataceae Trib. Sonerileae: Phyllagathis rajah and Sonerila metallica from Batang Ai, Sarawak, Borneo
FIGURE 1. Distribution map of Phyllagathis rajah and Sonerila metallica in Sarawak, Borneo (black circle).
AI Accountability Dataset
<p>The official data repository used for the AI Accountability paper by Samuel Lefcourt and Gregory Falco.</p>
Synthetic Datasets for "Binary Classification Optimisation with AI-Generated Data"
<p>Images of melanomas and Basal Cell Carcinoma generated with a stylegan2. Dataset corresponding to the article "Binary Classification Optimisation with AI-Generated Data"</p>
AI based essay grading survey
<p>Dataset of the survey on AI-based essay grading, conducted in <span>MCI | The Entrepreneurial School, Innsbruck, Austria. </span></p>
JuniperMapper: AI for Shrub Delineation (DATA)
<p>JuniperMapper is a dataset created to individually delineate <em>Juniperus</em> shrubs from satellite images using deep learning models. This data contains two types of datasets collected from the national park of Sierra Nevada, Spain:</p> <ol> <li>Photo Interpreted (PI) data: manually created by experts using a visual inspection of satellite images.</li> <li>Field Work (FW) data: manually created by experts using an in-situ inspection.</li> </ol> <p>Both datasets contain a set of RGB input images (.tif) at 13 cm resolution downloaded from Google Earth satellites, and their corresponding annotations (shapefiles) delineating each shrub. In addition, a .json file following the COCO dataset structure is also provided to train deep learning models.</p> <p>The PI dataset can be used to develop the deep learning model. It is already partitioned into Train, Validation, and Test sets. While the FW dataset can be used to perform the external validation of the model.</p>
Appendix A for the paper title : ArtInsight: A Multimodal AI Framework for Interpreting Children's Drawings and Enhancing Emotional Understanding
<p> Appendix A in this paper includes the specific prompt used for annotating the children's drawings with ChatGPT. This prompt was meticulously designed to ensure that the AI generated consistent and relevant descriptions and assessment reports for each image. The prompt guides the AI to focus on several key areas:</p> <ul> <li><strong>Description of the Artwork</strong>: Encouraging a detailed observation of the visual elements in the drawing, such as colors, shapes, figures, and any notable features.</li> <li><strong>Psychological Interpretation</strong>: Guiding the AI to interpret the possible emotional and psychological themes represented in the artwork based on the visual cues.</li> <li><strong>Possible Emotional Themes</strong>: Identifying underlying emotions or feelings that the child might be expressing through their drawing.</li> <li><strong>Recommended Interventions</strong>: Suggesting appropriate actions or conversations that therapists, educators, or parents could initiate to support the child's emotional well-being.</li> </ul>
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.