Collaboration Discovery, A Linked Data Approach

Investigators

  • Eric Rozell, Rensselaer Polytechnic Institute
  • Tom Narock, University of Maryland Baltimore County, tnarock@umbc.edu

Project Dates

  • January, 2013 thru Summer 2013

Project Description

Linked Data is a modern paradigm for data publishing, which leverages the inherent linking capabilities of the Web and provides semantics for the relationships between “things” on the Web. These “things” may be digital objects, such as publications and science data, or they may be representative of real-world entities, including people, organizations, and events. Recent Funding Friday projects and Summer Meeting sessions have demonstrated ESIP’s interest in Linked Data as well as modeling of organizations, people, and events that shape the ESIP network. At present, these two efforts have been pursued independently. We believe that the Linked Data paradigm is ideal for capturing and publishing information on the ESIP community and doing so would demonstrate the collaborative nature within ESIP.

This work will provide a scalable, open platform for current and future efforts to analyze and extend the community models in various ways. Through our previous efforts on a Funding Friday project, we have started to capture the ESIP research network as Linked Data. This project will further that effort by

  1. providing a collaborative development environment
  2. enabling a service that all of ESIP can access and
  3. extending our current Linked Data sources

Specifically, we will deploy a service on the ESIP Amazon Web Services that gives users the ability to:

  1. search for collaborators
  2. and discover relationships between ESIP and AGU/NSF projects

We will also focus on developing an ESIP platform for contributing to these datasets; we will seek out contributions from ESIP clusters which may include, but are not limited to

  1. interfaces for minor edits
  2. equivalence assertions
  3. batches uploads of new data

Current Status

We have been working with the AGU to convert past meeting data into Linked Open Data (LOD). These data have been linked to ESIP members and projects allowing cross-organizational queries. In addition, the AGU and ESIP LOD has been linked to the historical record of funded NSF projects. The AGU LOD is currently available http://198.61.161.98/abstracts/meetings Following final testing with AGU, the AGU interface will be rebranded abstracts.agu.org and publicly released. We anticipate a Fall 2013 release date.

The Amazon Cloud instance serving the NSF and ESIP LOD will be decommissioned at the conclusion of this project. The data will be transitioned to a public server at UMBC in Fall 2013. The data is also available upon request from the project investigators. All documentation regarding our experiences with Amazon Cloud and the creation of our data is available below.

Code

The code used in this project is available open source via our Google code site

Short name: 
esip_lod
Project type: 
Full project
Enable issue tracker: 
Yes