mdpi-ai/org-geo-ner

NER for affiliation parsing

org-geo-ner is a token-classification (NER) model that extracts organizations and geographic context from affiliation strings. It was fine-tuned in-house by MDPI AG to power mdpi-ror-search, an open-source tool that matches affiliation strings against the Research Organization Registry (ROR).

Model description

• Architecture: Fine-tuned from SIRIS-Lab/affilgood-affilxlm (AffilXLM), an XLM-RoBERTa base model (12 layers, 768 hidden size, ~270M parameters) further pre-trained with masked language modeling on 10 million OpenAlex affiliation strings (Duran-Silva et al., 2024)
• Task: Named Entity Recognition (NER) over affiliation / address strings
• Labeling scheme: BIO (B-/I-/O)
• Entity types:
  • ORG: parent organization (e.g. university, company, ministry)
  • SUB: sub-organization / department, faculty, institute, lab
  • CITY: city
  • COUNTRY: country
• Language: multilingual affiliation strings with a bias toward Latin-script, English-structured affiliation conventions
• License: Apache 2.0

Intended use

This model is designed to extract structured entities from free-text author affiliation strings as found in scholarly manuscripts and metadata (e.g. "Dept. of Physics, ETH, Zürich, Switzerland"). It is the first stage in a pipeline whose downstream goal is to resolve affiliations to ROR IDs. For further details about the pipeline, visit mdpi-ror-search.

It is not intended for general-purpose NER on arbitrary text (news, social media, etc.) as it is specialized for the affiliation/address domain.

How to use

The model works out of the box with the transformers token-classification pipeline using aggregation_strategy="simple", which merges wordpiece tokens into full entity spans.

from transformers import pipeline

ner_pipeline = pipeline(
    "token-classification",
    model="mdpi-ai/org-geo-ner",
    aggregation_strategy="simple",
)

text = "MDPI, Grosspeteranlage 5, 4052 Basel, Switzerland."
output = ner_pipeline(text)

for entity in output:
    print(f"{entity['entity_group']:<10} {entity['word']:<40} score={entity['score']:.3f}")

Output:

ORG             MDPI                           score=0.855
CITY            Basel                          score=0.994
COUNTRY         Switzerland                    score=1.000

Batch inference works the same way by passing a list of strings:

texts = [
    "MDPI, Grosspeteranlage 5, 4052 Basel, Switzerland.",
    "Department of Chemistry, MIT, Cambridge, MA, USA.",
]
outputs = ner_pipeline(texts)

For production use, filter low-confidence predictions with a score threshold (e.g. score >= 0.65) before downstream processing.

Training data

The model was fine-tuned on an internal, proprietary dataset of affiliation strings annotated for ORG, SUB, CITY, and COUNTRY spans. This dataset is not publicly released.

Evaluation

Evaluated with seqeval on the held-out test split (strict span matching):

Entity	Precision	Recall	F1
ORG	85.5	87.9	86.6
SUB	87.1	89.6	88.3
CITY	94.7	95.8	95.2
COUNTRY	97.8	98.8	98.3
Overall	90.5	92.3	91.4

Limitations and bias

• Trained on scholarly affiliation strings; performance may degrade on out-of-domain text.
• Street addresses, postal codes, and regions are not labeled and will typically be tagged O (e.g. MA in "Cambridge, MA, USA").
• The data is multilingual in organization and place names, but still skewed toward Latin-script, English-structured affiliation conventions; performance on affiliations written entirely in non-Latin scripts is not guaranteed.
• The training dataset is not publicly released, so exact reproduction of the reported metrics is not possible from this card alone.
• Ambiguous or heavily abbreviated affiliation strings (e.g. bare acronyms) may be mislabeled or missed; downstream consumers should apply a confidence threshold and treat predictions as candidates rather than ground truth.
• With aggregation_strategy="simple", hyphenated compound organization names (e.g. CREAF-CSIC-UAB) can be split into multiple adjacent ORG spans rather than merged into one, since the tokenizer treats - as a token boundary. Downstream consumers may want to re-merge adjacent same-label spans separated only by punctuation.

Additional information

• Repository (pipeline code, docs): MDPI-ROR-Search.
• Base model paper: Duran-Silva et al., 2024, AffilGood.
• Contact: For questions, issues, or collaboration inquiries, please open an issue on the GitHub repository.
• Funding: This project was fully funded and supported by MDPI AG.
• License: Apache 2.0
• Citation: If you use this model, please cite it as:

@misc{orggeoner,
  title  = {org-geo-ner: A Named Entity Recognition Model for Affiliation Parsing},
  author = {MDPI AG},
  year   = {2026},
  howpublished = {\url{https://huggingface.co/mdpi-ai/org-geo-ner}}
}

@inproceedings{duran-silva-etal-2024-affilgood,
  title     = {{A}ffil{G}ood: Building reliable institution name disambiguation tools to improve scientific literature analysis},
  author    = {Duran-Silva, Nicolau and Accuosto, Pablo and Przyby{\l}a, Piotr and Saggion, Horacio},
  booktitle = {Proceedings of the Fourth Workshop on Scholarly Document Processing (SDP 2024)},
  year      = {2024},
  address   = {Bangkok, Thailand},
  publisher = {Association for Computational Linguistics},
  url       = {https://aclanthology.org/2024.sdp-1.13},
  pages     = {135--144}
}

Disclaimer

This model is provided as-is, without warranty of any kind. Predictions should be reviewed before being used in contexts requiring high accuracy, such as automated compliance or reporting workflows.