Metadata Normalization in Scientific Articles

The metadata of a scientific article is its identity card in the open science ecosystem. An article with excellent content but inconsistent metadata — misspelled authors, missing DOI, generic affiliations — loses visibility in Google Scholar, Crossref, SciELO, and ORCID. Metadata normalization is the systematic process of correcting, standardizing, and enriching this data before publication.

What does "normalize" metadata mean?

Normalizing is not simply "cleaning" data. It involves transforming heterogeneous information — as editors receive it from authors — into structured records that comply with international standards:

Author name normalization

Authors submit names in diverse formats: "García, María Elena", "María Elena García", "M.E. García", "García M.E.". Without normalization, the same researcher appears as different authors in databases.

Practical rules

1. Separate surname from given name in <surname> and <given-names>.
2. Respect surname particles ("de la", "van", "del") as part of <surname>.
3. Do not abbreviate given names unless the journal policy explicitly allows it.
4. Preserve diacritical characters (accents, ñ, cedilla) in UTF-8.

<contrib contrib-type="author">
  <name>
    <surname>de la Rosa</surname>
    <given-names>María Fernanda</given-names>
  </name>
  <contrib-id contrib-id-type="orcid">https://orcid.org/0000-0003-4567-8901</contrib-id>
  <xref ref-type="aff" rid="aff1"><sup>1</sup></xref>
</contrib>

Corporate authors

For author groups (consortia, committees), use <collab> instead of <name>:

<contrib contrib-type="author">
  <collab>Public Health Research Group</collab>
</contrib>

ORCID: more than an identifier

ORCID links the researcher to all their publications persistently. ORCID normalization includes:

• Validate format: 16 digits in four groups separated by hyphens.
• Store full URI: https://orcid.org/0000-0001-2345-6789, not just the digits.
• Verify existence: query the public ORCID API to confirm the record is active.
• Do not invent ORCID: if the author does not provide one, omit the element; do not fill with placeholders.

ORCID also enables self-declaration of publications when the article includes the identifier correctly, improving author visibility without additional editor effort.

DOI: registration and tagging

The DOI is the persistent identifier that links the article to Crossref and facilitates citation. DOI normalization covers:

In the article (article-id)

<article-id pub-id-type="doi">10.1234/revista.2026.00123</article-id>

Without the https://doi.org/ prefix. No spaces or disallowed special characters.

In references (pub-id)

<pub-id pub-id-type="doi">10.5678/cited.2024.00456</pub-id>

Crossref registration

Before publishing, register the DOI in Crossref with complete metadata (title, authors, date, journal). A DOI registered late reduces early indexing in databases that harvest Crossref daily.

Institutional affiliations

Poorly normalized affiliations are a systemic problem. "Faculty of Medicine, University X" without country, city, or ROR identifier makes bibliometric analysis and research reporting difficult.

Recommended JATS structure

<aff id="aff1">
  <label>1</label>
  <institution content-type="orgname">Universidad Nacional Autónoma de México</institution>
  <institution content-type="orgdiv">Facultad de Ciencias</institution>
  <addr-line>Ciudad de México</addr-line>
  <country country="MX">México</country>
  <institution-id institution-id-type="ror">https://ror.org/01hcx6992</institution-id>
</aff>

Best practices

• Unify identical affiliations under a single id.
• Always include country, with ISO code in the country attribute.
• Consult ROR for standard institutional identifiers.
• Avoid non-standard abbreviations ("UNAM" without expanding at least once).

Editorial dates

Received, revised, accepted, and publication dates document the editorial cycle and are required by SciELO and PMC:

<history>
  <date date-type="received">
    <day>10</day><month>03</month><year>2026</year>
  </date>
  <date date-type="rev-rec">
    <day>25</day><month>04</month><year>2026</year>
  </date>
  <date date-type="accepted">
    <day>15</day><month>06</month><year>2026</year>
  </date>
</history>

Verify chronological coherence: received ≤ revision ≤ acceptance ≤ publication.

Keywords and classification

• Use xml:lang to indicate the language of each keyword group.
• Apply minimum and maximum per journal policy (typically 3–5).
• Consider controlled vocabularies (DeCS/MeSH for health sciences) when the journal is indexed in specialized databases.

<kwd-group kwd-group-type="author" xml:lang="en">
  <kwd>diabetes mellitus</kwd>
  <kwd>epidemiology</kwd>
  <kwd>Latin America</kwd>
</kwd-group>

Funding and conflicts of interest

Frequently overlooked metadata required by transparency policies:

<funding-group>
  <award-group>
    <funding-source>
      <institution-wrap>
        <institution>Consejo Nacional de Ciencia y Tecnología</institution>
      </institution-wrap>
    </funding-source>
    <award-id>PN-2024-12345</award-id>
  </award-group>
</funding-group>

And in the back matter:

<fn fn-type="COI-statement">
  <p>The authors declare no conflicts of interest.</p>
</fn>

Recommended normalization workflow

1. Collect metadata in a structured form at article acceptance.
2. Validate ORCID and reference DOIs via public APIs.
3. Normalize names and affiliations with documented rules.
4. Insert into JATS template with predefined mandatory fields.
5. Audit with an automatic validator before publication.

Run the JATS validator after inserting metadata to catch structural errors before delivery.

Impact of normalized metadata

Journals that implement systematic normalization report:

• Higher successful indexing rate on first SciELO submission.
• Better positioning in Google Scholar (authors linked correctly).
• Fewer author inquiries about publication visibility.
• Exportable data for institutional research reports.

Conclusion

Metadata normalization is not a cosmetic task after text editing: it is an editorial competency that determines whether published knowledge is discoverable, attributable, and reusable. Investing in normalization processes — author forms, ORCID validation, early DOI registration — multiplies the impact of every article the journal publishes.

For journals that need to scale this process without expanding staff, professional JATS XML production includes metadata normalization as a standard workflow stage. Request a quote to evaluate options for your publication volume.