美国NSF中的“图情”基金有哪些？(3)

Start Date:

June 1, 2017

End Date: [好文分享：www.11jj.com]

May 31, 2019 (Estimated)

Awarded Amount to Date:

$34,881.00

ABSTRACT
Digital data repositories have become increasingly important across researchcommunities in recent years due to changing expectations around data fromfunders and scientists. Managers, funders, and users of digital datarepositories are interested in assessing and communicating the impact of suchrepositories. A systematic approach and well-understood metrics, that may bequantitative or qualitative in nature, are needed to determine impact. Variousstakeholders that include repository evaluators, repository managers,information studies scholars, and researchers and other data producers need tocome together to develop relevant metrics to measure and document impact andoutcomes. To this end, the College of Information Studies, University ofMaryland, College Park proposes to facilitate a workshop that will gather thesevarious stakeholders for open discussion around digital data repository impactand evaluation. The goal of this workshop is to identify the current metrics,tools, and practices that are effective, and the issues that will requireadditional research.
Improvements to metrics used for the evaluation of digital repository servicesand their impacts on research is a topic which is well-suited to the workshopformat in this proposal. Conversations across stakeholder communities do nothappen often enough, but they are vital to the sustainable development ofdigital data repositories. This workshop will bring together a diverse set ofstakeholders around digital data repositories to discuss strategies and methodsto better assess the impact of repository services. Impact assessments are keyto understanding outcomes and necessary to refine and modify programs andinfrastructures. The development of more robust repository metrics willinfluence and transform practices in these organizations, broadening access toresearch data. Additionally, a report will be publicly released following theworkshop so the results will have increased visibility beyond the participantsand their institutions.

CRI: CI-SUSTAIN: Collaborative Research: CiteSeerX:Toward Sustainable Support of Scholarly Big Data

Program Manager:

Wendy Nilsen
 CNS Division Of Computer and Network Systems
 CSE Direct For Computer & Info Scie & Enginr

Start Date:

August 1, 2018

End Date:

October 31, 2018 (Estimated)

Awarded Amount to Date:

$230,000.00

ABSTRACT
Access to the scientific and scholarly literature has changed radically inrecent decades. Increasingly researchers and scholars make their publicationsfreely available on the Web. Taking advantage of this opportunity, newscientific search engine tools have been developed such as Google Scholar,Semantic Scholar, and CiteSeer, now CiteSeerX. CiteSeerX has become one of themost comprehensive and widely-used online public resources for the Computer andInformation Science and Engineering (CISE) research community. Millions ofCiteSeerX Portable Document Format (PDF) documents are indexed by Google.CiteSeerX is unique among digital library search engines. It is open access,most all of its documents are harvested from the public Web, and users havefull-text access to all documents searchable on its website. Moreover, itprovides all automatically extracted metadata and citation context via an OpenArchive Initiative (OAI) metadata service interface and bulk downloads on apublic cloud - all under a Creative Commons license. This service is usuallynot available from other scholarly search engines. CiteSeerX performs automaticextraction and indexing of tables (in production), figures (developed)}, andalgorithms (developed), capabilities rarely seen in other scholarly searchengines. CiteSeerX provides its open source software and architecture onGitHub. At this time none of the other above-mentioned systems release theirdigital library software. 
Utilizing the established CiteSeerX infrastructure, this proposal aims tocreate a sustainable CiteSeerX system with new data resources and a much largerdata collection. We will develop a new system that runs with low operationoverhead, without a single point of failure, and that provides quality andenriched data and metadata in portable formats that will be available throughaccessible user interfaces. We will ingest all freely accessible scientificdocuments on the Web, currently estimated to be 30 million. CiteSeerX will makeavailable high-quality metadata through an accessible Web User Interface,Application Programming Interface, and data dumps. SeerSuite, the platform onwhich CiteSeerX is built, will be refactored so as to be an easily deployableand configurable scholarly digital library framework. It will be built oncommercial grade open source software. In addition, we will provide searchablesemantic metadata, such as key phrases and disambiguated author names, andnon-textual content such as data from figures, tables, algorithms, andequations. For long-term sustainability we will explore different monetizationmodels. The result will be a refactored digital library search engine thatprovides stable, usable, and reliable data services on multiple types ofscientific documents built on a portable, maintainable, and self-containedframework that can be deployed for other research document digital collections.Source code will be hosted at https://github.com/SeerLabs. System developmentand related research will be published in relevant venues and be made publiclyavailable.

 

CyberTraining: CIU: Preparing the Public Sector ResearchWorkforce to Impact Communities through Data Science

Program Manager:

Sushil Prasad
 OAC Office of Advanced Cyberinfrastructure (OAC)
 CSE Direct For Computer & Info Scie & Enginr

Start Date: