Linking research articles and datasets using ORCID profiles

2014-01-24

By Maurice Vanderfeesten

The basic idea is to allow researchers to indicate what datasetsare used in what publications, that both are listed on their ORCID profile.

Context:

  1. ORCID is a service that allows individual researcher to claim an “Open Researcher and Contributor Identifier”. ( see , create profile, login, continue reading)
  2. ORCID allows researchers to make a Profile page with a list of their academic work.
  3. Within your ORCID-profile you can add your academic work manually or automatically.
  4. Automatic import of your academic work is done by searching for your name variances and/or your ORCID in various metadata sources(DOI registrars). The current metadata sources populate publications and datasets (e.g. Crossef, DataCite, etc.)

Problem:

  1. The problem is that there is no problem.
  2. Moving Publications and Datasets so close to each other on one profile page of the author, begs for making a simple link between the two object types.

Solution:

  1. Create a service that makes this link possible.

Effect:

  1. The effect it will bring is a network of datasets pointing to different publications and vice versa[1].
  2. Large amounts of value will be created for ranking lists, citation analysis and visualisation services.

Incentive:

  1. The incentive for researchers doing this linking effortis citation. Research Data is nog yet a first class citizen in research output, but it will be in the near future.

Operation:

  1. Create a user friendly ‘enhancing service’ that links datasets and publications.
  2. This could be done by ORCID, but the ORCID API allows third parties to create this type of services where researchers can authorize the third party to access their profile with the list of works.

Requirements:

  1. ORCID should keep track of the ‘type’ of Work (data or publications), and provide this to the third party ‘enhancing service’. This allows the service to easily offer separated lists of work.
  2. Preferably the ‘enhancing service’ should send the enhanced information to a central node, creating the biggest value, creating an umbrella for all ‘enhanced services’ that might pop-up. This bulk of relation data allows others services to use this information in ranking, citation network analysis and visualisations more easy, instead of getting the information from all 3rd party services again.

[1] Preferable ontology for semantic web inferences: or inverse