Academic Track Record is the primary source for establishing trust between collaborators that don't know each other.
Because science is universal, global and involves mobility, these encounters occur very often.

In such events, the researchers are left to check to past affiliations of each other, look for collaborators they shared, see what impactful conference or journal paper the other appeared in, see if the other supervised or reviewed PhDs, postgrads in relevant topics. Hence, a semi-formalized trust chain in established.

In order to establish more trust in a researcher account in an academic collaborations, there are several automated actions an AAI platform can take. Commercial (Academia.edu, researchergate, google scholar) and community-owned (ORCID) initiatives already perform very basic collection of information (scraping crossref metadata (DOI)-s and the web). These methods could be much enhanced with more assured information that we have in the Research and Education space and could enrich an institutional or a  MyAccessID account, for example.

Several parts of this concept has been proven and demonstrated by the various science social networks, like Academia.edu and ResearchGate, who, as soon as a publication appears with a DOI. This is done by regularly scraping the related database, and the same happens for citations. This very often happens with matching of name strings, in lack of better curated attributes in the crossref metadata and results in mis-attributed data. However, other, equally important elements of the record - peer reviews in and efforts service of science, like PhD defense committee membership, and altmetrics (contribution to research software, instruments; confirmed reader counts) are overlooked and the technology for that is only an idea at this moment.

A) arXiv API+ORCID: in possession of a verified ORCID, the arXiv API can be queried for articles written by an author:

https://arxiv.org/search/advanced?terms-0-operator=AND&terms-0-term=&terms-0-field=title&terms-1-operator=AND&terms-1-term=0000-0002-9979-9101&terms-1-field=orcid&classification-physics_archives=all&classification-include_cross_list=include&date-filter_by=all_dates&date-year=&date-from_date=&date-to_date=&date-date_type=submitted_date&abstracts=show&size=50&order=-announced_date_first

Trust: high

arXiv was originally created for physics and is still dominant on that field.

Output DOI+publishing place

B) Crossref API+ORCID

In the crossref JSON metadata, ORCID is present, if it was known

{"ORCID":"http:\/\/orcid.org\/0000-0002-9979-9101","authenticated-orcid":false,"given":"Mih\u00e1ly","family":"H\u00e9der","sequence":"additional","affiliation":[]}]

C) DBLP+ORCID

on DBLP is possible to search by ORCID

D) email based matching

E) name based matching

trust: low

F) Consuming Verifiable Credentials

The main use case is when an editor, committee, a funding organization encounters a researcher they don't know.  In such events, the researchers are left to check to past affiliations of each other, look for collaborators they shared, see what impactful conference or journal paper the other appeared in, see if the other supervised or reviewed PhDs, postgrads in relevant topics. Hence, a semi-formalized trust chain in established.

The team members will attempt methods A-E as described in the original topic description. Additionally, RoR db and edugain metadata should be investigated as a source.

The possible outcomes if this activity are the following:

• OJS (open journal system) integration PoC
• command-line toolkit
• article, best practices

The following parties will use the results of this activity:

The following results were created and delivered:

OJS Integration with mock data:

OJS details