EstBERT_NER Open - source Named Entity Recognition Model: Accurately Identify Three Types of Entities in Estonian Texts

Estbert NER

Developed by tartuNLP

A named entity recognition model fine-tuned based on EstBERT, used to identify three types of entities in Estonian texts: locations, organizations, and personal names.

Sequence Labeling Other#Estonian NER #BERT fine-tuning #Entity recognition

Downloads 516

Release Time : 3/2/2022

Model Overview

This model is a named entity recognition model based on the EstBERT architecture, specifically designed for processing Estonian texts. It can identify three types of entities: locations (LOC), organizations (ORG), and personal names (PER).

Model Features

Specialized for Estonian

A named entity recognition model optimized specifically for the Estonian language.

Three-Type Entity Recognition

Capable of accurately identifying three types of entities: locations (LOC), organizations (ORG), and personal names (PER).

Fine-Tuned Based on EstBERT

Utilizes the pre-trained EstBERT model for fine-tuning, offering better language understanding capabilities.

Model Capabilities

Estonian text processing

Named entity recognition

Entity classification

Use Cases

Information Extraction

News Analysis

Extracting key entity information from Estonian news articles.

Automatically identifies people, organizations, and locations mentioned in the news.

Financial Document Processing

Analyzing organization names and transaction information in financial documents.

Successfully identifies key information such as bank names in examples.

🚀 EstBERT_NER

EstBERT_NER is a fine - tuned EstBERT model designed for Named Entity Recognition. It can identify key entities in Estonian text, offering practical solutions for text analysis.

🚀 Quick Start

EstBERT_NER is a fine - tuned EstBERT model that can be used for Named Entity Recognition. This model was trained on the Estonian NER dataset created by Tkachenko et al. It can recognize three types of entities: locations (LOC), organizations (ORG) and persons (PER).

💻 Usage Examples

Basic Usage

# You can use this model with Transformers pipeline for NER. Post-processing of results may be necessary as the model occasionally tags subword tokens as entities. 
from transformers import BertTokenizer, BertForTokenClassification
from transformers import pipeline

tokenizer = BertTokenizer.from_pretrained('tartuNLP/EstBERT_NER')
bertner = BertForTokenClassification.from_pretrained('tartuNLP/EstBERT_NER')

nlp = pipeline("ner", model=bertner, tokenizer=tokenizer)
sentence = 'Eesti Ekspressi teada on Eesti Pank uurinud Hansapanga tehinguid , mis toimusid kaks aastat tagasi suvel ja mille käigus voolas panka ligi miljardi krooni ulatuses kahtlast raha .'

ner_results = nlp(sentence)
print(ner_results)

Advanced Usage

# The output shows the recognized entities and their corresponding scores.
[{'word': 'Eesti', 'score': 0.9964128136634827, 'entity': 'B-ORG', 'index': 1}, {'word': 'Ekspressi', 'score': 0.9978809356689453, 'entity': 'I-ORG', 'index': 2}, {'word': 'Eesti', 'score': 0.9988121390342712, 'entity': 'B-ORG', 'index': 5}, {'word': 'Pank', 'score': 0.9985784292221069, 'entity': 'I-ORG', 'index': 6}, {'word': 'Hansapanga', 'score': 0.9979034662246704, 'entity': 'B-ORG', 'index': 8}]

📄 License

This project is licensed under the CC BY 4.0 license.

📚 Documentation

BibTeX entry and citation info

@misc{tanvir2020estbert,
      title={EstBERT: A Pretrained Language-Specific BERT for Estonian}, 
      author={Hasan Tanvir and Claudia Kittask and Kairit Sirts},
      year={2020},
      eprint={2011.04784},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}

Property	Details
Base Model	tartuNLP/EstBERT
License	CC BY 4.0

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご