Search NASA⌕ Search

SEARCH · Search NASA

Results for “Dataset Curation”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 127 records · Page 7
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Reference Site Conditions for Floating Wind Arrays in the United States

Floating offshore wind farm design is highly site-specific, requiring detailed information about the specific conditions of a project area for realistic design studies. Unfortunately, publicly available site condition data for potential floating offshore wind project sites in the United States is scarce. To support U.S. offshore wind research, we developed reference site condition datasets, including metocean and seabed information, for four potential floating wind project areas in the U.S.: Humboldt Bay, Morro Bay, the Gulf of Maine, and the Gulf of Mexico. These datasets were compiled using publicly available data. Our metocean analysis, covering wind, waves, and surface currents, utilized measurement data from 2000 to 2020. Sources included the National Renewable Energy Laboratory’s National Offshore Wind Dataset for wind data, National Data Buoy Center buoys for wave data, and the High Frequency Radar Network for surface currents. These data were integrated into hourly time series used to compute extreme return periods up to 500 years, monthly statistics, and joint probability clusters for fatigue analysis. Soil conditions were evaluated using the usSEABED database and bathymetry grids were interpolated from the NCEI Digital Elevation Model Global Mosaic. In addition to providing curated reference site condition datasets for four U.S. areas, our assessment highlights the need for more publicly available metocean and soil condition data.

17 WIND ENERGY↗
DOE OSTI2026

AI-Ready Data Pilot Project Report

The proliferation of artificial intelligence in scientific research has created an urgent need to define "AI-ready data" for researchers and, more importantly, provide resources to help them produce AI-ready data. At Pacific Northwest National Laboratory, we conducted a pilot study with three data scientists evaluating three CSV datasets from different scientific domains, followed by semi-structured interviews capturing assessment practices. Our findings reveal that AI-readiness evaluation is intuition-based, with practitioners asking "How fast can I go from raw data to my machine learning pipeline?" Data scientists consistently prioritized workflow efficiency, human interpretability, and quality stewardship signals. From these insights, we developed a practical evaluation framework comprising data requirements, metadata standards, and validation tests that provides actionable criteria for producing and curating AI-ready datasets, addressing the gap between theoretical understanding and practical implementation.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Active learning for the design of polycrystalline textures using conditional normalizing flows

Generative modeling has opened new avenues for solving previously intractable materials design problems. However, these new opportunities are accompanied by a drastic increase in the required amount of training data. This is in stark juxtaposition to the high expense and difficulty in curating such large materials datasets. In this work, we propose a novel framework for integrating generative models within an active learning loop. Further, this enables the training of generative models with datasets significantly smaller than what has previously been demonstrated, providing a direct route for their application in data constrained environments. The functionality of this framework is then demonstrated by addressing the challenge of designing polycrystalline textures associated with target anisotropic mechanical properties. The developed protocol exhibited a cost reduction between 14 to 18 times over a randomly sampled experimental design.

36 MATERIALS SCIENCE↗
DOE OSTI2024

Physics-coupled data-driven design of high-temperature alloys

We present a materials design loop, which streamlines physics-coupled machine learning (ML) surrogate models to discover new alloy chemistries with improved properties. The efficacy is demonstrated by discovering a high-temperature alumina-forming austenitic (AFA) stainless steel with enhanced creep, followed by experimental validation. The ML models have been trained using a well-curated, highly consistent experimental dataset augmented with synthetic microstructural features from a computational thermodynamic approach. We have populated a large number of hypothetical AFA alloys to explore the high-dimensional composition space and have predicted their creep properties by providing the same synthetic input features obtained from the trained ML models. Uncertainties from the ML training were taken as thresholds for truncating predicted results to identify alloys with improved or deteriorated creep. Individual elemental compositions have been determined via probability density distribution analysis from the group of alloys at the top and bottom of the predicted creep values for further virtual and experimental validations. In conclusion, we anticipate that this workflow can be applied to screen desired conditions, such as chemistry and processing parameters, in high-dimensional space through physics-guided data analytics.

Alloy design↗
DOE OSTI2026

Agentic framework for programmatic crystal structure generation using a fine-tuned worker–supervisor large language model

Platinum group metals (PGMs) underpin many catalytic technologies but face severe supply constraints, motivating the search for alternative materials and computational methods to accelerate discovery. While atomistic simulation tools such as Pymatgen and ASE have streamlined structure manipulation, they require detailed inputs, limiting accessibility for experimentalists and slowing early-stage exploration. Here, in this study, we present an AI-driven agentic framework that orchestrates worker–supervisor large language models (LLMs). The worker translates natural-language prompts of varying abstraction into valid crystallographic structures using a compact LLM fine-tuned with low-rank adaptation on a curated text–code–CIF dataset, emphasizing energy-efficient training. Benchmarking against the baseline CodeGen-350M-mono model shows that fine-tuning reduces hallucination rates from 100% to as low as 5% and improves structural match accuracy to up to 82% for fully specified inputs. Accuracy declines with decreasing prompt detail but remains nontrivial even when only stoichiometry and space group are provided, underscoring the LLM’s capacity for crystallographic inference. The supervisor Claude LLM evaluates the outputs and triggers iterative refinement through the worker’s built-in structure manipulation capabilities (e.g., supercell scaling, strain, vacancy, and substitution operations). We further demonstrate use cases for technologically relevant catalysts, including IrO 2 , pyrochlore Pb 2 Ir 2 O 7 , Ni 2 FeO 4 , and Ni 3 Mo, where the framework generates physically consistent structures that can be refined via geometry optimization. This work introduces a low-energy, language-driven pathway for integrating human and machine intelligence in materials design, paving the way for AI-assisted synthesis planning and high-throughput screening of complex oxides.

AI agent↗
DOE OSTI2025

DeepBench: A simulation package for physical benchmarking data

We introduce **DeepBench**, a python library that generates simple simulated image data from first principles, such as basic geometric shapes and astronomical objects. These data are highly valuable for developing (calibration, testing, and benchmarking) statistical and machine learning models because they make it possible to connect the final data product to physically interpretable inputs. This software includes tools to curate and store the datasets to maximize reproducibility.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2022

Identifying Disinformation Using Rhetorical Devices in Natural Language Models

Foreign disinformation campaigns are strategically organized, extended efforts using disinformation – false or misleading information deliberately placed by an adversary – to achieve some goal. Disinformation campaigns pose severe threats to our nation’s security by misinforming decision makers and negatively influencing their actions when they are operating on limited amounts of evidence. Current efforts rely on subject matter experts to manually identify disinformation, or on computers and traditional natural language processing algorithms to identify patterns in data to calculate the probability that something is disinformation or not. While both have their merits and successes, subject matter experts are unable to keep up with the high volumes of global information and traditional natural language algorithms do not do well in identifying “why” something is disinformation or not. Our hypothesis is that we can identify disinformation by looking at the way someone speaks, in the rhetorical devices they use. We have curated and annotated a dataset designed for multiple natural language processing tasks, but specifically useful for disinformation detection algorithms.

97 MATHEMATICS AND COMPUTING↗
NASA NTRSDate not supplied

Enabling API Access to the Space Weather Services at the Community Coordinated Modeling Center

Over the span of 20 years, the Community Coordinated Modeling Center (CCMC, https://ccmc.gsfc.nasa.gov) has been leading a number of community-driven services and applications that provide a free and open access to the cutting-edge space weather and Heliophysics models through a simple web-based interface. CCMC also oversees an open archive of user model simulations and related metadata, maintains space weather-related data streams, curates validation and event datasets, and more. To maximize utility of its complementary services and data holdings, CCMC has been gradually building up an ad-hoc set of interfaces and specifications that facilitate coupling and interconnection within the organization, while also simplifying management and monitoring of the data. As the models continue to grow in maturity and complexity, CCMC is looking to reduce the complexity for its end users by making public some of the internal APIs as well as implementing dedicated interfaces as required by the community. In this presentation, we will overview the current run services at CCMC and will describe our current and near-future efforts in providing interfaces to these services.

space weather↗
DOE OSTI2024

Towards Diverse and Representative Global Pretraining Datasets for Remote Sensing Foundation Models

The design of a pretraining dataset is emerging as a critical component for the generality of foundation models. In the remote sensing realm, large volumes of imagery and benchmark datasets exist that can be leveraged to pretrain foundation models, however using this imagery in absence of a well-crafted sampling strategy is inefficient and has the potential to create biased and less generalizable models. Here, we provide a discussion and vision for the curation and assessment of pretraining datasets for remote sensing geospatial foundation models. We highlight the importance of geographic, temporal, and image acquisition diversity and review possible strategies to enable such diversity at global scale. In addition to these characteristics, support for various spatial-temporal pretext tasks within the dataset is also critical. Ultimately, our primary objective is to place emphasis on and draw attention to the data curation stage of the foundation model development pipeline. By doing so, we think it is possible to reduce biases of geospatial foundation models, as well as enable broader generalization to downstream remote sensing tasks and applications.

Arndt, Jacob↗
DOE OSTI2023

Expanding Accurate Person Recognition to New Altitudes and Ranges: The BRIAR Dataset

Face recognition technology has advanced significantly in recent years due largely to the availability of large and increasingly complex training datasets for use in deep learning models. These datasets, however, typically comprise images scraped from news sites or social media plat-forms and, therefore, have limited utility in more advanced security, forensics, and military applications. These applications require lower resolution, longer ranges, and ele-vated viewpoints. To meet these critical needs, we collected and curated the first and second subsets of a large multi-modal biometric dataset designed for use in the research and development (R&D) of biometric recognition technolo-gies under extremely challenging conditions. Thus far, the dataset includes more than 350,000 still images and over 1,300 hours of video footage of approximately 1,000 sub-jects. To collect this data, we used Nikon DSLR cameras, a variety of commercial surveillance cameras, specialized long-rage R&D cameras, and Group 1 and Group 2 UAV platforms. The goal is to support the development of algorithms capable of accurately recognizing people at ranges up to 1,000 m and from high angles of elevation. These ad-vances will include improvements to the state of the art in face recognition and will support new research in the area of whole-body recognition using methods based on gait and anthropometry. This paper describes methods used to col-lect and curate the dataset, and the dataset's characteristics at the current stage.

Brogan, Joel↗
NASA NTRS2017

New GES DISC Services Shortening the Path in Science Data Discovery

The Current GES DISC available services only allow user to select variables from a single dataset at a time and too many variables from a dataset are displayed, choice is hard. At American Geophysical Union (AGU) 2016 Fall Meeting, Goddard Earth Sciences Data Information Services Center (GES DISC) unveiled a new service: Datalist. A Datalist is a collection of predefined or user-defined data variables from one or more archived datasets. Our science support team curated predefined datalist and provided value to the user community. Imagine some novice user wants to study hurricane and typed in hurricane in the search box. The first item in the search result is GES DISC provided Hurricane Datalist. It contains scientists recommended variables from multiple datasets like TRMM, GPM, MERRA, etc. Datalist uses the same architecture as that of our new website, which also provides one-stop shopping for data, metadata, citation, documentation, visualization and other available services.We implemented Datalist with new GES DISC web architecture, one single web page that unified all user interfaces. From that webpage, users can find data by either type in keyword, or browse by category. It also provides user with a sophisticated integrated data and services package, including metadata, citation, documentation, visualization, and data-specific services, all available from one-stop shopping.

Datalist↗
DOE OSTI2022

CTSA MHRI Datasets

Oak Ridge National Laboratory (ORNL) has collaborated with MedStar Health Research Institute (MHRI) to develop, test, and validate health outcomes using electronic health records (EHRs) from hospitals associated with participating Clinical and Translational Science Awards (CTSA). MHRI, the research organization of MedStar Health (MSH), has a history of initiating projects, both in the laboratory and in the field, that serve the needs of medically underserved and disenfranchised groups. In this document, the ORNL team is providing data curation documentation for the following datasets, which are publicly available: 1. Area Deprivation Index (ADI) 2015 block group; 2. Child Opportunity Index 2015 tract; 3. Low food access 2017 block group; 4. Neighborhood deprivation index 2017 tract; 5. Social Capital Index 2014 county; and, 6. Social Vulnerability Index 2014 tract.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Recommendations for developing, documenting, and distributing data products derived from NEON data

The National Ecological Observatory Network (NEON) provides over 180 distinct data products from 81 sites (47 terrestrial and 34 freshwater aquatic sites) within the United States and Puerto Rico. These data products include both field and remote sensing data collected using standardized protocols and sampling schema, with centralized quality assurance and quality control (QA/QC) provided by NEON staff. Such breadth of data creates opportunities for the research community to extend basic and applied research while also extending the impact and reach of NEON data through the creation of derived data products—higher level data products derived by the user community from NEON data. Derived data products are curated, documented, reproducibly-generated datasets created by applying various processing steps to one or more lower level data products—including interpolation, extrapolation, integration, statistical analysis, modeling, or transformations. Derived data products directly benefit the research community and increase the impact of NEON data by broadening the size and diversity of the user base, decreasing the time and effort needed for working with NEON data, providing primary research foci through the development via the derivation process, and helping users address multidisciplinary questions. Creating derived data products also promotes personal career advancement to those involved through publications, citations, and future grant proposals. However, the creation of derived data products is a nontrivial task. Here we provide an overview of the process of creating derived data products while outlining the advantages, challenges, and major considerations.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2026

Status on Microstructural Studies of Nuclear Graphite

This report documents the completion of the Advanced Reactor Technologies (ART) Level 3 Milestone (M3AT-26OR0605054), “Provide status on microstructural studies of nuclear graphite,” due May 1, 2026. The contents of this report summarize recent progress in the development of the Nuclear Graphite Microstructure Library, including datasets that have been generated, curated, and submitted for publication, as part of the broader effort to establish and integrate the library within the NDMAS platform.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
NASA NTRSDate not supplied

Integrating Multi-agency Data Products in a Cloud-based Platform for Streamlined Discovery, Visualization, and Use

Earth science data users almost always have an interest in utilizing geospatial data from multiple agencies. As computing capability and cloud-based infrastructures accelerate the pace at which scientific research can be done, there is a growing need to enable search, discovery, and use of multi-agency geospatial observations relevant for a common use case - without undergoing the search and discovery process in a less efficient, disparate path with each agency. NASA’s Earth Observing System Data and Information System (EOSDIS) and NOAA’s National Environmental Satellite, Data and Information Service (NESDIS) both support a wide range of Earth science disciplines’ research, operations, and applications activities. Presently, however, there are few examples of data discovery frameworks supporting an inquiry of both NASA’s and NOAA’s extensive archives of Earth observations that are equally suitable for a particular science scenario, regardless of the agency that “owns” the data. NASA and NOAA are collaborating on a data expedition platform for exploring fire weather using data products from both agencies. Users will be able to search, discover, and visualize NASA and NOAA products in one interface. Each agency will curate metadata for its respective datasets, providing for a rich search experience. The collaboration will pilot a shared search interface into these metadata datastores. Data products will be stored in the cloud in cloud-optimized format(s). These formats will allow for optimized data access and visualization to support the “data expedition”. Avenues for further development and application of this cloud-based, multi-agency data provisioning platform will also be discussed.

cloud-based technology↗
NASA NTRSDate not supplied

Senteniel-6 Radio Occultation Product Released by NASA GES DISC to Supplement Satellite Remote Sensing Datasets for PBL Study

The NASA Goddard Earth Sciences Data and Information Services Center (GES DISC) curates hyperspectral atmospheric sounder remote sensing and numerical model reanalysis datasets which have been utilized in the Planetary Boundary Layer (PBL) research and applications. The hyperspectral sounder remote-sensing datasets include the Atmospheric Infrared Sounder (AIRS) on the Aqua satellite to the Cross-track Infrared Sounder (CrIS) on Suomi--National Polar- orbiting Partnership (NPP) and National Oceanic and Atmospheric Administration -20 (N NOAA-20)/ Joint Polar-orbiting Satellite System -1 (JPSS-1). The Modern-Era Retrospective analysis for Research and Applications Version 2 (MERRA-2) global reanalysis product provides a data record commencing in 1980. The sounder remote sensing and reanalysis datasets include temperature, water vapor, and trace gas profile down to the PBL, and also have a derived PBL height as well. A nearly 10-year (June 2006 to December 2015) seasonal and annual PBL height climatology dataset from COSMIC Global Navigation Satellite System (GNSS) radio occultation (RO) measurement is also available from the GES DISC. In collaboration with Sentinel-6 Project, the GES DISC is implementing curation activities for GNSS RO products from the Sentinel-6A/Sentinel-6 Michael Freilich satellite launched on November 21, 2020. Sentinel-6A RO products provide refractivity, temperature, and humidity profile with finer vertical resolution, leveraging PBL research and application as a supplement to the hyperspectral sounder remote sensing and reanalysis products. The public release of Sentinel- 6A RO products is scheduled for mid-October of 2021. In this presentation, we will introduce all Senitnel-6A products and services, and demonstrate use cases studying the PBL by combining these products with other GES DISC archived data products.

Feng Ding↗
NASA NTRSDate not supplied

Expanding Biological Repository Data Available for Sharing and Knowledge Discovery

Biology has developed next-generation data science and alternative analytical approaches with methodologies which require principal investigator (PI) experimental assay data be re-used. This new approach involves mining multiple datasets at once from various hierarchical organizations of biological complexity, while concurrently evaluating how experimental factors affect endpoints of standard assays. The purpose of the NASA Ames Life Sciences Data Archive (ALSDA) is to collect, curate, and make findable, accessible, interoperable, and reusable (FAIR) all non-human space-relevant biological data. These data include mission metadata, subject metadata, assay metadata (parameters), raw and processed assay data, assay imagery, and subject-experienced telemetry (radiation, temperature, humidity, acoustics, vibrations). ALSDA has transformed to bring current biological repository data and all future collected data into this new scientific data mining reality. It has integrated into the ‘NASA Open Science’ group of projects to facilitate a suite of new tools and workflows to improve data accessibility and reusability by implementing data management plans, automating data submission agreements, and adopting the single-point-of-entry data submission portal, originally developed by NASA GeneLab. These systems required ALSDA to develop science assay configurations for the submission portal, capturing essential assay parameters according to established norms in each sub-field within biology. The submission portal expedites data collection by enhancing ease of PI data submission, providing a user interface and specificity for which data is to be submitted. ALSDA datasets are curated to maintain rich metadata, accuracy of datasets, data transparency, provenance, and additionally ensure data are machine-readable (e.g., R and Python languages). ALSDA integration with GeneLab and its analysis portals enable higher-order physiological-level datasets be mined in conjunction with -omics datasets. As ALSDA physiological-level datasets are published (micro-computed tomography, histology, intraocular pressure, hormonal assays, immunostaining, ultrasonography), the merging of hierarchical organizations of biological complexity from spaceflight will enable new knowledge discovery approaches.

Ryan T Scott↗
DOE OSTI2026

Knowledge-guided learning with curated prior genetic biomarkers for robust model interpretation

Abstract Motivation Knowledge-guided learning offers effective and robust model training strategies in data-scarce settings by incorporating established domain knowledge, thereby enhancing generalization, robustness, and interpretability. By contrast, conventional deep learning approaches rely purely on data-driven learning, which can limit robust model interpretability, particularly in high-dimensional settings with limited size samples. In computational biology, knowledge-guided learning has primarily leveraged network- and structural-based knowledge, leading to biologically interpretable representations and enhanced predictive performance compared to conventional approaches. However, curated biomarkers, one of the most accessible forms of biological knowledge, remain largely unexplored within knowledge-guided paradigms. Results In this study, we propose a model-agnostic training paradigm, Biomarker-driven Explainable Prior-guided Learning (BioExPL), that can be applied to any neural networks that incorporates curated prior knowledge. BioExPL enforces neural networks to reflect curated biomarker priors in their latent representations through a novel knowledge-alignment loss. BioExPL consistently demonstrated significantly improved predictive performance and enhanced model interpretability with minimized computational overhead in simulation studies and intensive experiments on multiple cancer datasets. BioExPL not only integrates prior curated knowledge into the model but also accurately identifies unknown associated signals additionally. BioExPL is model-agnostic and domain-independent, enabling its integration into diverse neural network architectures. Availability and implementation The open-source is publicly available at: https://github.com/datax-lab/BioExPL.

Baek, Beomsu [Department of Computer Science, Univ↗