DOAJ
This section describes how the OpenAIRE Graph integrates the metadata of the articles indexed by the Directory of Open Access Journals (DOAJ), using the DOAJ data dump. It covers where the data comes from, how it is acquired and transformed, and how the DOAJ metadata fields populate the fields of the OpenAIRE Graph data model.
Data source
DOAJ is a community-curated directory of open access, peer-reviewed journals. Besides the journals, it indexes the metadata of their articles. The journal and article metadata are released under a CC0 waiver.
OpenAIRE collects the DOAJ articles for the data source DOAJ. These records used to be collected through the DOAJ OAI-PMH interface; since August 2026 they are collected from the data dump. The records keep the identifiers they had before, so the dump-based collection replaces the previous one record by record.
Data acquisition
DOAJ publishes full dumps of its article metadata and of its journal metadata in JSON format, generated every month. Access to the dumps is granted by DOAJ on request. The article dump of July 2026 contained 13,301,270 articles; the journal dump contained 23,247 journals.
Every collection imports a complete dump, so the DOAJ records in the Graph reflect the dump available at collection time.
Processing workflow
- Download: the article dump and the journal dump are downloaded from DOAJ.
- Conversion: every article becomes one XML record in Dublin Core, reproducing the record DOAJ published through OAI-PMH, with the same identifier (
oai:doaj.org/article:<id>). The volume, issue and pages are written as a citation (<journal title>, Vol <volume>, Iss <issue>, Pp <first page>-<last page>), as DOAJ did in OAI-PMH. Two values that the OAI-PMH records did not carry are added: the ORCID iD of the authors, and the licence of the journal, taken from the journal dump by ISSN. The few characters that XML 1.0 does not allow are removed. - Collection: the OpenAIRE aggregation system collects the resulting file for the data source DOAJ, one record per article.
- Transformation: the transformation rule
xslt_dc2oaf_DOAJ_dump_hadoopmaps each record into the OpenAIRE internal metadata format. The mapping choices described below are made at this step. - Graph production: each transformed record becomes a research product, which then goes through the standard steps of the Graph production workflow (cleaning, deduplication, enrichment).
Record filtering
No article is filtered out: every article of the dump becomes a research product.
Entity Mapping
The table below describes how the fields of a DOAJ article populate the fields of an OpenAIRE research product.
JSON paths refer to an article of the DOAJ article dump; [] marks a list.
| OpenAIRE Research Product field path | DOAJ record JSON path | Notes |
|---|---|---|
| Research product | ||
id | id | id in the form doajarticles::md5(oai:doaj.org/article:<id>), built from the local identifier, because DOAJ is not a PID authority |
originalIds | id | oai:doaj.org/article:<id> and the internal identifier |
type | (fixed value) | publication: DOAJ indexes journal articles only |
mainTitle | bibjson.title | |
authors | bibjson.author[] | One author per entry, in the order of the source |
descriptions | bibjson.abstract | |
publicationDate | bibjson.year, bibjson.month | The first day of the month, or of the year when the month is missing |
publisher | bibjson.journal.publisher | |
language | bibjson.journal.language[] | The first language of the journal (DOAJ gives no language for the article), cleaned by using vocabulary dnet:languages |
subjects | bibjson.subject[].term, bibjson.subject[].code, bibjson.keywords[] | scheme=keyword. The subject terms and codes come from the Library of Congress Classification assigned by DOAJ to the journal |
sources | bibjson.journal.title, bibjson.journal.volume, bibjson.journal.number, bibjson.start_page, bibjson.end_page | The citation built by the conversion, e.g. IEEE Access, Vol 12, Pp 11010-11023 |
bestAccessRight | (fixed value) | OPEN, from instances.accessRight |
collectedfrom | (fixed value) | The data source DOAJ, driver______::bee53aa31dc2cbb538c10c2b65fa5824. The same data source is set as the host of the instance |
| Author Mapping | ||
authors.fullName | bibjson.author[].name | The name as given by the source |
authors.name | bibjson.author[].name | Derived from the full name by the Graph, when it can be split |
authors.surname | bibjson.author[].name | Derived from the full name by the Graph, when it can be split |
authors.rank | Sequential number starting from 1 | |
authors.pid | bibjson.author[].orcid_id | scheme = orcid_pending (about 6% of the articles have at least one ORCID iD) |
| Container Mapping | ||
container.name | bibjson.journal.title | |
container.issnPrinted | bibjson.identifier[] | The first ISSN of the article (type pissn or eissn), whether print or electronic |
container.vol | bibjson.journal.volume | Numeric values only |
container.iss | bibjson.journal.number | Numeric values only |
container.sp | bibjson.start_page | Only when both pages are present and numeric |
container.ep | bibjson.end_page | Only when both pages are present and numeric |
| Instance Mapping | One instance per record | |
instances.type | (fixed value) | Article, from the dnet:publication_resource vocabulary |
instances.alternateIdentifiers | bibjson.identifier[] | The DOI (type = doi, about 82% of the articles), with scheme = doi. DOAJ is not a PID authority for DOIs, so the DOI is an alternate identifier, not a value of instances.pids |
instances.urls | bibjson.link[], id | The full-text links (type = fulltext) and the DOAJ page of the article, https://doaj.org/article/<id> |
instances.accessRight | (fixed value) | OPEN: DOAJ indexes open access journals only |
instances.license | bibjson.license[].url of the journal | The licence of the journal, taken from the journal dump by ISSN and cleaned by using vocabulary dnet:licenses (Creative Commons and public domain licences, about 95% of the articles); other licences, such as a publisher's own licence, are not mapped |
instances.publicationDate | bibjson.year, bibjson.month | As publicationDate |
instances.refereed | (fixed value) | peerReviewed by default: DOAJ indexes journals that apply peer review or editorial quality control |
All the records are related to the data source with name=DOAJ and id=driver______::bee53aa31dc2cbb538c10c2b65fa5824.
Relation Mapping
The DOAJ records do not create relations: the dump has no references and no funding information, and the affiliations of the authors have no identifiers.
Fields not mapped
The following DOAJ fields do not populate the Graph: the affiliations of the authors (bibjson.author[].affiliation), the country of the journal (bibjson.journal.country), the content type of the full-text links (bibjson.link[].content_type), and the record dates (created_date, last_updated).