NomadLite

From TRAP-RCUB

Revision as of 21:33, 20 September 2021 by Trap (talk | contribs)

NomadLite is a service developed by the TRAP-RCUB team. Each repository developed by TRAP-RCUB is supplied with an independent instance of NomadLite. The source code is not open and the service is avaiable to repository managers (log in is required) from the institutions having a TRAP-RCUB repository.

NomadLite is a complex service relying on text mining and a set of public APIs. once checked and verified by repository managers, the retrieved information is automatically inserted into appropriate metadata fields;

NomanLite has the following functionalities:

  • Retrieving funding information based on mining the full text of deposited documents. A list of potential matches is displayed to the repository manager. The repository manager selects the correct option, if available in the list and, once confirmed, this information is automatically inserted into the dc.relation field. In case the text in a document is not readable (no OCR, corrupred files, inappropriate encoding, etc.), NomadLite will not be able to process it and no information will be retrieved. In this case, the repository manager can add the finding information manually. NomadLite searches for a limited number of funding lines: EU FP7, Horizon 2020, Horizon Europe; projects funded by the Serbian Ministry of Education, Science and Technological Development (2006-2010 and 2011-2019), institutional funding by MESTD since 2020, and the projects funded by the Fund for Science of the Republic of Serbia (since 2020). Funding information related to other funding lines can be added manually in the appropriate metadata field in the repository.
  • Assigning Scopus and Web of Science identifiers. NomadLite matches metadata records from the repository and Scopus and Web of Science relying on APIs. A list of potential matches is displayed to the repository manager. Whene the repository manager confirms a match is correct, Scopus / WoS identifier is inserted into the appropriate field in the repository. Thanks to this, it is possible to display citation counts from Scopus and WoS on the landing pages of publications.
  • Assigning ranking information to journal articles according to the national research evaluation system). This feature relies on the ProRef service, developed by the University of Belgrade Computer Centre. Among other things, ProRef offers information about journal rankings according to the national research evaluation system. All records in the repository that contain information in the dc.identifier.issn field are checked against the datbase underlying ProRef. If the ISSN is found, a repository manager will see a list of options to chose from. Upon checking the correct option, the ranking information will be inserted in the appropriate custom dield in the repository. This feature is not available in DAIS and RIFDT.
  • Retrieving available Open Access versions based on CORE Discovery and the Unpaywall API. Potential matches are displayed to the repository manager and they have to be checked manually to establish whether the retrieved version is appropriate. In case the version is correct, the submission to the repository is conducted either via the submission form or through Ellena.