AffilGood-NER-multilingual

Overview

Click to expand

• Model type: Language Model
• Architecture: XLM-RoBERTa-base
• Language: Multilingual
• License: Apache 2.0
• Task: Named Entity Recognition
• Data: AffilGood-NER
• Additional Resources:
  • Paper
  • GitHub

Model description

The multilingual version of affilgood-NER-multilingual is a Named Entity Recognition (NER) model for identifying named entities in raw affiliation strings from scientific papers and projects, fine-tuned from the AffilXLM model, a XLM-RoBERTa base model futher pre-trained for MLM task on a medium-size corpus of raw affiliation stirngs collected from OpenAlex.

It has been trained with a dataset that contains 7 main types of entities from multilingual raw affiliation strings texts, with 5,266 texts.

After analyzing hundreds of affiliations from multiple countries and languages, we defined seven entity types: SUB-ORGANISATION, ORGANISATION, CITY, COUNTRY, ADDRESS, POSTAL-CODE, and REGION, detailed [annotation guidelines here].

Identifying named entities (organization names, cities, countries) in affiliation strings not only enables more effective linking with external organization registries, but it can also play an essential role in the geolocation of organizations and can also contribute to identify organizations and their position in an institutional hierarchy -- especially for those not listed in external databases. Information automatically extracted by means of a NER model can also facilitate the construction of knowledge graphs, and support the development of manually curated registries.

Intended Usage

This model is intended to be used for multilingual raw affiliation strings, because this model is pre-trained on XLM-RoBERTa, NER and large further pre-training corpora are both multilingual.

How to use

from transformers import pipeline
affilgood_ner_pipeline("ner", model=model, tokenizer=tokenizer, aggregation_strategy="simple")
sentence = "CSIC, Global ecology Unit CREAF-CSIC-UAB, Bellaterra 08193, Catalonia, Spain."
output = affilgood_ner_pipeline(sentence)
print(output)

Limitations and bias

No measures have been taken to estimate the bias and toxicity embedded in the model.

The NER dataset contains 5,266 raw affiliation strings obtained from OpenAlex. It includes multilingual samples from all available countries and geographies to ensure comprehensive coverage and diversity. To enable our model to recognize various affiliation string formats, the dataset includes a wide range of structures, different ways of grouping main and subsidiary institutions and various methods of separating organization names. We also included ill-formed affiliations and those containing errors resulting from automatic extraction from PDF files.

Training

We used the AffilGood-NER dataset for training and evaluation.

We fine-tuned the adapted and base models for token classification with the IOB annotation schema. We trained the models for 25 epochs, using 80% of the dataset for training, 10% for validation and 10% for testing.

Hyperparameters used for training are described here:

• Learning Rate: 2e-5
• Learning Rate Decay: Linear
• Weight Decay: 0.01
• Warmup Portion: 0.06
• Batch Size: 128
• Number of Steps: 25k steps
• Adam ε: 1e-6
• Adam β₁: 0.9
• Adam β₂: 0.999

The best performing epoch (considering macro-averaged F1 with strict matching criteria) was used to select the model.

Evaluation

The model's performance was evaluated on a 10% of the dataset.

Category	RoBERTa	XLM	AffilRoBERTa	AffilXLM (this model)
ALL	.910	.915	.920	.925
-----	------	------	------	----------
ORG	.869	.886	.879	.906
SUB	.898	.890	.911	.892
CITY	.936	.941	.950	.958
COUNTRY	.971	.973	.980	.970
REGION	.870	.876	.874	.882
POSTAL	.975	.975	.981	.966
ADDRESS	.804	.811	.794	.869

All the numbers reported above represent F1-score with strict match, when both the boundaries and types of the entities match.

Additional information

Authors

• SIRIS Lab, Research Division of SIRIS Academic, Barcelona, Spain
• LaSTUS Lab, TALN Group, Universitat Pompeu Fabra, Barcelona, Spain
• Institute of Computer Science, Polish Academy of Sciences, Warsaw, Poland

Contact

For further information, send an email to either nicolau.duransilva@sirisacademic.com or info@sirisacademic.com.

License

This work is distributed under a Apache License, Version 2.0.

Funding

This work was partially funded and supporter by:

• Industrial Doctorates Plan of the Department of Research and Universities of the Generalitat de Catalunya, by Departament de Recerca i Universitats de la Generalitat de Catalunya (ajuts SGR-Cat 2021),
• Maria de Maeztu Units of Excellence Programme CEX2021-001195-M, funded by MCIN/AEI /10.13039/501100011033
• EU HORIZON SciLake (Grant Agreement 101058573)
• EU HORIZON ERINIA (Grant Agreement 101060930)

Citation

@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",
    editor = "Ghosal, Tirthankar  and
      Singh, Amanpreet  and
      Waard, Anita  and
      Mayr, Philipp  and
      Naik, Aakanksha  and
      Weller, Orion  and
      Lee, Yoonjoo  and
      Shen, Shannon  and
      Qin, Yanxia",
    booktitle = "Proceedings of the Fourth Workshop on Scholarly Document Processing (SDP 2024)",
    month = aug,
    year = "2024",
    address = "Bangkok, Thailand",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2024.sdp-1.13",
    pages = "135--144",
}

Disclaimer

Click to expand

The model published in this repository is intended for a generalist purpose and is made available to third parties under a Apache v2.0 License.

Please keep in mind that the model may have bias and/or any other undesirable distortions. When third parties deploy or provide systems and/or services to other parties using this model (or a system based on it) or become users of the model itself, they should note that it is under their responsibility to mitigate the risks arising from its use and, in any event, to comply with applicable regulations, including regulations regarding the use of Artificial Intelligence.

In no event shall the owners and creators of the model be liable for any results arising from the use made by third parties.