Wikineural Multilingual Ner

🚀 WikiNEuRal: Combined Neural and Knowledge-based Silver Data Creation for Multilingual NER

This project fine - tunes a multilingual language model (mBERT) on the WikiNEuRal dataset for Named Entity Recognition (NER), supporting 9 languages.

🚀 Quick Start

WikiNEuRal is a multilingual NER project. We fine - tuned a multilingual language model (mBERT) for 3 epochs on the WikiNEuRal dataset for Named Entity Recognition (NER). The resulting model supports 9 languages (de, en, es, fr, it, nl, pl, pt, ru) and was trained on all these languages jointly.

✨ Features

• Multilingual Support: Supports 9 languages including German, English, Spanish, French, Italian, Dutch, Polish, Portuguese, and Russian.
• Combined Approach: Combines neural and knowledge - based methods for silver data creation.
• High - Quality Data: Uses Wikipedia texts and novel techniques to produce high - quality training corpora for NER.

📦 Installation

No specific installation steps are provided in the original README.

💻 Usage Examples

Basic Usage

from transformers import AutoTokenizer, AutoModelForTokenClassification
from transformers import pipeline

tokenizer = AutoTokenizer.from_pretrained("Babelscape/wikineural-multilingual-ner")
model = AutoModelForTokenClassification.from_pretrained("Babelscape/wikineural-multilingual-ner")

nlp = pipeline("ner", model=model, tokenizer=tokenizer, grouped_entities=True)
example = "My name is Wolfgang and I live in Berlin"

ner_results = nlp(example)
print(ner_results)

📚 Documentation

This is the model card for the EMNLP 2021 paper WikiNEuRal: Combined Neural and Knowledge-based Silver Data Creation for Multilingual NER.

Citation: If you use the model, please reference this work in your paper:

@inproceedings{tedeschi-etal-2021-wikineural-combined,
    title = "{W}iki{NE}u{R}al: {C}ombined Neural and Knowledge-based Silver Data Creation for Multilingual {NER}",
    author = "Tedeschi, Simone  and
      Maiorca, Valentino  and
      Campolungo, Niccol{\`o}  and
      Cecconi, Francesco  and
      Navigli, Roberto",
    booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2021",
    month = nov,
    year = "2021",
    address = "Punta Cana, Dominican Republic",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2021.findings-emnlp.215",
    pages = "2521--2533",
    abstract = "Multilingual Named Entity Recognition (NER) is a key intermediate task which is needed in many areas of NLP. In this paper, we address the well-known issue of data scarcity in NER, especially relevant when moving to a multilingual scenario, and go beyond current approaches to the creation of multilingual silver data for the task. We exploit the texts of Wikipedia and introduce a new methodology based on the effective combination of knowledge-based approaches and neural models, together with a novel domain adaptation technique, to produce high-quality training corpora for NER. We evaluate our datasets extensively on standard benchmarks for NER, yielding substantial improvements up to 6 span-based F1-score points over previous state-of-the-art systems for data creation.",
}

The original repository for the paper can be found at https://github.com/Babelscape/wikineural.

🔧 Technical Details

The model is a fine - tuned multilingual language model (mBERT) for 3 epochs on the WikiNEuRal dataset. It addresses the data scarcity issue in multilingual NER by combining knowledge - based approaches and neural models, along with a novel domain adaptation technique.

📄 License

Contents of this repository are restricted to only non - commercial research purposes under the [Creative Commons Attribution - NonCommercial - ShareAlike 4.0 International License (CC BY - NC - SA 4.0)](https://creativecommons.org/licenses/by - nc - sa/4.0/). Copyright of the dataset contents and models belongs to the original copyright holders.

⚠️ Important Note

This model is trained on WikiNEuRal, a state - of - the - art dataset for Multilingual NER automatically derived from Wikipedia. Therefore, it might not generalize well to all textual genres (e.g. news). On the other hand, models trained only on news articles (e.g. only on CoNLL03) have been proven to obtain much lower scores on encyclopedic articles. To obtain more robust systems, we encourage you to train a system on the combination of WikiNEuRal with other datasets (e.g. WikiNEuRal + CoNLL).