ner-distilbert-base Open-source English Named Entity Recognition Model - Free Deployment for Precise Entity Recognition

Ner Distilbert Base Uncased Ontonotesv5 Englishv4

Developed by djagatiya

Named entity recognition model based on distilbert-base-uncased architecture, fine-tuned on the conll2012_ontonotesv5-english-v4 dataset

Sequence Labeling

Transformers

#High-precision NER #Multi-category entity recognition #English text processing

Downloads 18

Release Time : 7/3/2022

Model Overview

This model is used to identify named entities in text, supporting recognition of 18 entity types including persons, locations, organizations, etc.

Model Features

High-precision entity recognition

Achieves an F1 score of 85.53 across 18 entity types

Lightweight model

Based on DistilBERT architecture, reducing model size while maintaining performance

Broad entity coverage

Supports recognition of 18 different types of named entities

Model Capabilities

Text entity recognition

Multi-category entity classification

Natural language processing

Use Cases

Information extraction

News text analysis

Extract key information such as persons, organizations, and locations from news articles

Accurately identifies key entities in news

Document processing

Automatically identify relevant entities in legal documents or contracts

Helps quickly locate key information in documents

Knowledge graph construction

Knowledge graph entity extraction

Extract entities from unstructured text for knowledge graph construction

Provides structured data sources for knowledge graphs

🚀 (NER) distilbert-base-uncased : conll2012_ontonotesv5-english-v4

This distilbert-base-uncased NER model was fine - tuned on the conll2012_ontonotesv5-english-v4 dataset. It offers an effective solution for named - entity recognition tasks. Check out NER - System Repository for more information.

🚀 Quick Start

The model is ready to use after fine - tuning on the specified dataset. You can refer to the NER - System Repository for detailed usage instructions.

✨ Features

Trained on the conll2012_ontonotesv5-english-v4 dataset, which provides a solid foundation for English NER tasks.
Achieves good evaluation metrics, indicating its effectiveness in real - world scenarios.

📚 Documentation

🔍 Evaluation

Precision: 84.60
Recall: 86.47
F1 - Score: 85.53

⚠️ Important Note

Check out this eval.log file for evaluation metrics and classification report.

               precision    recall  f1-score   support

    CARDINAL       0.84      0.86      0.85       935
        DATE       0.83      0.88      0.85      1602
       EVENT       0.57      0.57      0.57        63
         FAC       0.55      0.62      0.58       135
         GPE       0.95      0.92      0.94      2240
    LANGUAGE       0.82      0.64      0.72        22
         LAW       0.50      0.50      0.50        40
         LOC       0.55      0.72      0.62       179
       MONEY       0.87      0.89      0.88       314
        NORP       0.85      0.89      0.87       841
     ORDINAL       0.81      0.88      0.84       195
         ORG       0.81      0.83      0.82      1795
     PERCENT       0.87      0.89      0.88       349
      PERSON       0.93      0.93      0.93      1988
     PRODUCT       0.55      0.55      0.55        76
    QUANTITY       0.71      0.80      0.75       105
        TIME       0.59      0.66      0.62       212
 WORK_OF_ART       0.42      0.44      0.43       166

   micro avg       0.85      0.86      0.86     11257
   macro avg       0.72      0.75      0.73     11257
weighted avg       0.85      0.86      0.86     11257

Property	Details
Tags	token - classification
Datasets	djagatiya/ner - ontonotes - v5 - eng - v4

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