Search NASA⌕ Search

SEARCH · Search NASA

Results for “knowledge capture”

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 19 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2013

Incremental learning for automated knowledge capture

People responding to high-consequence national-security situations need tools to help them make the right decision quickly. The dynamic, time-critical, and ever-changing nature of these situations, especially those involving an adversary, require models of decision support that can dynamically react as a situation unfolds and changes. Automated knowledge capture is a key part of creating individualized models of decision making in many situations because it has been demonstrated as a very robust way to populate computational models of cognition. However, existing automated knowledge capture techniques only populate a knowledge model with data prior to its use, after which the knowledge model is static and unchanging. In contrast, humans, including our national-security adversaries, continually learn, adapt, and create new knowledge as they make decisions and witness their effect. This artificial dichotomy between creation and use exists because the majority of automated knowledge capture techniques are based on traditional batch machine-learning and statistical algorithms. These algorithms are primarily designed to optimize the accuracy of their predictions and only secondarily, if at all, concerned with issues such as speed, memory use, or ability to be incrementally updated. Thus, when new data arrives, batch algorithms used for automated knowledge capture currently require significant recomputation, frequently from scratch, which makes them ill suited for use in dynamic, timecritical, high-consequence decision making environments. In this work we seek to explore and expand upon the capabilities of dynamic, incremental models that can adapt to an ever-changing feature space.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Enhancing Knowledge Capture and Management within a Knowledge-Driven Workforce [Capstone Project]

This project aimed to understand how the collective knowledge of a group of engineers or ‘knowledge-based workers’ can be baselined through use of qualitative questionnaires based on input from subject matter experts. The results of the questionnaires were then used to inform management and the group where to focus efforts on terms of training and mentoring. With use of re-administering the qualitative questionnaire to gauge progress over time. Additionally, the development a knowledge-based wiki was explored to help shape group culture from one where ideas and knowledge tend to be siloed, into one where information is traded freely across the group and with external customers and collaborators.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗
DOE OSTI2021

Workshop on Using a Git-based Repository for ASTM Committee Technical Knowledge Capture [Slides]

Consensus technical standards maintained by ASTM International rely on individual contributors to act as technical contacts, effectively stewards, for a particular standard. As a particular standard’s technical content evolves over time, the contact must update relevant tables, graphics, and supporting discussion. However, the details of the data sources and/or processes used to perform such updates may be separate enough from the standard such that expert knowledge is required to perform these updates. As a result of this need for expert knowledge, and the desire to capture that knowledge, subcommittee E10.05 (Nuclear Radiation Metrology) has experimented with using a git-based repository to capture data and processes to reproduce figures needed by E10.05 to maintain the ASTM standards under their jurisdiction. This workshop will provide an overview of the git-based repository site’s organization, operation, and outputs. It will then lead interested workshop participants through adding content, not yet already included in the repository to the site.

42 ENGINEERING↗
DOE OSTI2023

A Detonator Design Guide and Knowledge Capture System [Slides]

Engineering design--good design--is a blend of engineering and science; this is especially true for of explosive detonator design. Detonators are complex components with interrelated mechanical, electrical, and chemical attributes, all of which must be balanced to achieve a robust and reliable, working whole. This design guide presents an integrated, lifecycle approach to the design process, bringing together engineering design with science, emphasizing elegance and subtlety in the design process.

42 ENGINEERING↗
DOE OSTI2019

Unsupervised word embeddings capture latent knowledge from materials science literature

The overwhelming majority of scientific knowledge is published as text, which is difficult to analyse by either traditional statistical analysis or modern machine learning methods. By contrast, the main source of machine-interpretable data for the materials research community has come from structured property databases, which encompass only a small fraction of the knowledge present in the research literature. Beyond property values, publications contain valuable knowledge regarding the connections and relationships between data items as interpreted by the authors. To improve the identification and use of this knowledge, several studies have focused on the retrieval of information from scientific literature using supervised natural language processing, which requires large hand-labelled datasets for training. Here we show that materials science knowledge present in the published literature can be efficiently encoded as information-dense word embeddings (vector representations of words) without human labelling or supervision. Without any explicit insertion of chemical knowledge, these embeddings capture complex materials science concepts such as the underlying structure of the periodic table and structure-property relationships in materials. Furthermore, we demonstrate that an unsupervised method can recommend materials for functional applications several years before their discovery. This suggests that latent knowledge regarding future discoveries is to a large extent embedded in past publications. Lastly, our findings highlight the possibility of extracting knowledge and relationships from the massive body of scientific literature in a collective manner, and point towards a generalized approach to the mining of scientific literature.

36 MATERIALS SCIENCE↗
DOE OSTI2015

WE-F-BRB-01: The Power of Ontologies and Standardized Terminologies for Capturing Clinical Knowledge

Advancements in informatics in radiotherapy are opening up opportunities to improve our ability to assess treatment plans. Models on individualizing patient dose constraints from prior patient data and shape relationships have been extensively researched and are now making their way into commercial products. New developments in knowledge based treatment planning involve understanding the impact of the radiation dosimetry on the patient. Akin to radiobiology models that have driven intensity modulated radiotherapy optimization, toxicity and outcome predictions based on treatment plans and prior patient experiences may be the next step in knowledge based planning. In order to realize these predictions, it is necessary to understand how the clinical information can be captured, structured and organized with ontologies and databases designed for recall. Large databases containing radiation dosimetry and outcomes present the opportunity to evaluate treatment plans against predictions of toxicity and disease response. Such evaluations can be based on dose volume histogram or even the full 3-dimensional dose distribution and its relation to the critical anatomy. This session will provide an understanding of ontologies and standard terminologies used to capture clinical knowledge into structured databases; How data can be organized and accessed to utilize the knowledge in planning; and examples of research and clinical efforts to incorporate that clinical knowledge into planning for improved care for our patients. Learning Objectives: Understand the role of standard terminologies, ontologies and data organization in oncology Understand methods to capture clinical toxicity and outcomes in a clinical setting Understand opportunities to learn from clinical data and its application to treatment planning Todd McNutt receives funding from Philips, Elekta and Toshiba for some of the work presented.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2021

Supervised learning with word embeddings derived from PubMed captures latent knowledge about protein kinases and cancer

Abstract Inhibiting protein kinases (PKs) that cause cancers has been an important topic in cancer therapy for years. So far, almost 8% of >530 PKs have been targeted by FDA-approved medications, and around 150 protein kinase inhibitors (PKIs) have been tested in clinical trials. We present an approach based on natural language processing and machine learning to investigate the relations between PKs and cancers, predicting PKs whose inhibition would be efficacious to treat a certain cancer. Our approach represents PKs and cancers as semantically meaningful 100-dimensional vectors based on word and concept neighborhoods in PubMed abstracts. We use information about phase I-IV trials in ClinicalTrials.gov to construct a training set for random forest classification. Our results with historical data show that associations between PKs and specific cancers can be predicted years in advance with good accuracy. Our tool can be used to predict the relevance of inhibiting PKs for specific cancers and to support the design of well-focused clinical trials to discover novel PKIs for cancer therapy.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2016

Knowledge represented using RDF semantic network in the concept of semantic web

The RDF(S) model has been declared as the basic model to capture knowledge of the semantic web. It provides a common and flexible way to decompose composed knowledge to elementary statements, which can be represented by RDF triples or by RDF graph vectors. From the logical point of view, elements of knowledge can be expressed using at most binary predicates, which can be converted to RDF-triples or graph vectors. However, it is not able to capture implicit knowledge representable by logical formulas. This contribution shows how existing approaches (semantic networks and clausal form logic) can be combined together with RDF to obtain RDF-compatible system with ability to represent implicit knowledge and inference over knowledge base.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

Knowledge Management in Nuclear Liabilities Regulation in the UK - 18155

Within the UK's nuclear safety regulator (Office for Nuclear Regulation; ONR) nuclear liabilities regulation addresses the safe management of radioactive waste, decommissioning and land quality through a facility's lifetime. Higher activity radioactive waste (HAW) could be stored on site for up to 150 years prior to disposal, Magnox decommissioning sites are planned to be in Care and Maintenance for up to 70 years and Sellafield decommissioning is not planned to be complete until after 2100. The long timescales over which decommissioning occurs and radioactive waste is managed is a challenge to maintaining a suitable level of knowledge to effectively and efficiently regulate nuclear liabilities activities. There is a danger that implicit knowledge accumulated over decades could be lost as people retire. Effective knowledge capture and transfer are key elements in maintaining a workforce with the capacity and capability to deliver high levels of nuclear safety, given the demographics of the UK nuclear industry. Maintenance of recorded information over long timescales with changing technologies is necessary to allow safe storage and disposal of waste and intelligent response to incidents. As the safety regulator, ONR's ability to influence improvement in nuclear safety relies on our inspectors having access to high quality knowledge. The ONR Strategy 2016-2020 [1,2] commits to continue to recruit, retain and develop staff so that we have a core workforce with the capacity and capability to deliver high quality, enabling nuclear regulation. Over the next 30 years, the UK plans to develop a geological disposal facility and UK EDF power stations will transition into decommissioning. The ONR Knowledge Management Strategy sets out the activities planned to maintain specialist knowledge to support staff development and build organisational resilience. Guidance on waste management is given in 'The management of higher activity radioactive waste on nuclear licensed sites' (referred to as the Joint Guidance [3]) published by ONR and the UK environmental regulators. It includes our expectations for managing information and records relating to radioactive waste, and introduces the concept of Radioactive Waste Management Cases (RWMCs). Our expectation is that knowledge and records must both be managed and transferred in such a way that future licensees are properly equipped to ensure that waste continues to be managed in a manner that prevents or minimises hazards to human health and the environment. RWMCs are an effective vehicle for demonstrating how the key elements of long-term safety and environmental performance will be delivered over the life cycle of the waste. They give a history of how the waste has been managed, capturing information and change over time. This paper will look at the challenges to managing knowledge in the UK nuclear industry, focusing on the safe management of radioactive waste, decommissioning and land quality. It will look at the problems faced by both licensees and ONR, drawing out common challenges. The paper will outline the approach taken within ONR to capture and communicate waste management and decommissioning knowledge and experience. Finally, it will look at how ONR regulates and influences how knowledge and records are dealt with in the industry to ensure safe nuclear liabilities management in the long term. (authors)

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2020

Emerging materials intelligence ecosystems propelled by machine learning

We report that the age of cognitive computing and artificial intelligence (AI) is just dawning. Inspired by its successes and promises, several AI ecosystems are blossoming, many of them within the domain of materials science and engineering. These materials intelligence ecosystems are being shaped by several independent developments. Machine learning (ML) algorithms and extant materials data are utilized to create surrogate models of materials properties and performance predictions. Materials data repositories, which fuel such surrogate model development, are mushrooming. Automated data and knowledge capture from the literature (to populate data repositories) using natural language processing approaches is being explored. The design of materials that meet target property requirements and of synthesis steps to create target materials appear to be within reach, either by closed-loop active-learning strategies or by inverting the prediction pipeline using advanced generative algorithms. AI and ML concepts are also transforming the computational and physical laboratory infrastructural landscapes used to create materials data in the first place. Surrogate models that can outstrip physics-based simulations (on which they are trained) by several orders of magnitude in speed while preserving accuracy are being actively developed. Automation, autonomy and guided high-throughput techniques are imparting enormous efficiencies and eliminating redundancies in materials synthesis and characterization. The integration of the various parts of the burgeoning ML landscape may lead to materials-savvy digital assistants and to a human-machine partnership that could enable dramatic efficiencies, accelerated discoveries and increased productivity. Here, we review these emergent materials intelligence ecosystems and discuss the imminent challenges and opportunities. The materials research landscape is being transformed by the infusion of approaches based on machine learning. This Review discusses the emerging materials intelligence ecosystems and the potential of human-machine partnerships for fast and efficient virtual materials screening, development and discovery.

36 MATERIALS SCIENCE↗
DOE OSTI2014

Sandia Cyber Omni Tracker

SCOT cyber security team enhancement tool that coordinates activities, captures knowledge, and serves as a platform to automate time-consuming tasks that a cyber security team needs to perform in its daily operations.

Nick Peterson, ToddBruner↗
DOE OSTI2021

MenuFramework

The Menu Framework is a reusable Extended Reality multi-level menu system for the Unity Development Platform. The menu structure is architected for reusability to suit whatever need the developer may have for a multi-level, reusable menu framework. Menu Framework's architecture, GUI, and functionality are based on the paper "Extended Reality for Enhanced Training and Knowledge Capture" (SAND2020-2361C). This tool uses the Microsoft Mixed Reality ToolKit (MRTK) built in functionality to provide a suite of functionalities including bounding box, follow me, minimize, maximize, settings, and help commands. Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525. SAND2021-1507 O

Klein, BrandonThorin↗
DOE OSTI2016

T.Rex

T.Rex is used to explore tabular data sets containing up to ten million records to help rapidly understand a previously unknown data set. Analysis can quickly identify patterns of interest and the records and fields that capture those patterns. T.Rex contains a growing set of deep analytical tools and supports robust export capabilities that selected data can be incorporated into to other specialized tools for further analysis. T.Rex is flexible in ingesting different types and formats of data, allowing the user to interactively experiment and perform trial and error guesses on the structure of the data; and also has a variety of linked visual analytic tools that enable exploration of the data to find relevant content, relationships among content, trends within the content, and capture knowledge about the content. Finally, T.Rex has a rich export capability, to extract relevant subsets of a larger data source, to further analyze their data in other analytic tools.

Source record↗
DOE OSTI2012

Automated Expert Modeling and Student Evaluation

AEMASE searches a database of recorded events for combinations of events that are of interest. It compares matching combinations to a statistical model to determine similarity to previous events of interest and alerts the user as new matching examples are found. AEMASE is currently used by weapons tactics instructors to find situations of interest in recorded tactical training scenarios. AEMASE builds on a sub-component, the Relational Blackboard (RBB), which is being released as open-source software. AEMASE builds on RBB adding interactive expert model construction (automated knowledge capture) and re-evaluation of scenario data.

Source record↗
DOE OSTI2014

Sandia Cyber Omni Tracker

SCOT cyber security team enhancement tool that coordinates activities, captures knowledge, and serves as a platform to automate time-consuming tasks that a cyber security team needs to perform in its daily operations.

Nick Peterson, Todd Bruner↗
DOE OSTI2022

2.3.3.404 - National Lab and University Collaboration for MHK Instrumentation and Data Processing Tools

Field and laboratory validation, testing, demonstration, and operation are critical steps for increasing the technology readiness level of marine energy (ME) converters because they provide high-quality testing and performance data that are critical information used to feed all aspects of technology development. This project, in partnership with industry, enables the marine and hydrokinetic energy (MHK) community to reliably and efficiently collect, process, manage, and share quality data by facilitating access to and development of instrumentation, guidelines and data processing/QA tools. Under this project, open-source data processing code (MHKiT) and tools (ME Data Pipeline, MRE Code Hub, PRIMRE Code Catalog), instrumentation (loads measurements), data acquisition systems (miniDAQ), and measurement guidance tools (Telesto, high EMI guidance) were developed to facilitate the collection and processing of quality laboratory and field data. Overall, this project is intended to improve the quality of the data collected during laboratory and field demonstration projects by standardizing the collection and processing techniques, as well as by improving access to instrumentation, code, and measurement guidance. Quality data will, in turn, lead to improved knowledge capture following ME device testing.

data processing↗