albert-fa-zwnj-base-v2-ner Open-Source Model - Persian Named Entity Recognition Supporting 10 Entity Types

Home

Albert Fa Zwnj Base V2 Ner

Developed by HooshvareLab

Albert model fine-tuned for Persian named entity recognition task, supporting 10 entity types

Sequence Labeling

Transformers

Other#Persian Named Entity Recognition #Multi-dataset Fine-tuning #Ten-class Entity Annotation

Downloads 57

Release Time : 3/2/2022

Model Overview

This is a Persian named entity recognition model based on Albert architecture, fine-tuned on hybrid NER datasets, capable of recognizing 10 entity types including dates, events, facilities, locations, etc.

Model Features

Multi-entity Type Recognition

Supports recognition of 10 different named entity types including dates, events, locations, persons, etc.

High-precision Performance

Achieves an overall F1 score of 0.9414 on test sets, with most entity types exceeding 0.9 F1 score

Hybrid Dataset Training

Trained using three Persian NER datasets: ARMAN, PEYMA and WikiANN

Model Capabilities

Persian text analysis

Named Entity Recognition

Multi-category entity classification

Use Cases

Text Analysis

News Text Analysis

Extract key information like persons, organizations, locations from Persian news

Can accurately identify key entities in news

Social Media Monitoring

Analyze entity information in Persian social media content

Helps track trending persons, events and organizations

Information Extraction

Document Automation Processing

Automatically extract structured information from Persian documents

Improves document processing efficiency

🚀 AlbertNER

This model is fine-tuned for the Named Entity Recognition (NER) task. It uses a mixed NER dataset collected from ARMAN, PEYMA, and WikiANN. The dataset covers ten types of entities:

Date (DAT)
Event (EVE)
Facility (FAC)
Location (LOC)
Money (MON)
Organization (ORG)
Percent (PCT)
Person (PER)
Product (PRO)
Time (TIM)

✨ Features

Fine-tuned for NER on a diverse dataset.
Covers ten common types of entities.

📦 Installation

Installing requirements

pip install sentencepiece
pip install transformers

💻 Usage Examples

Basic Usage

from transformers import AutoTokenizer
from transformers import AutoModelForTokenClassification  # for pytorch
from transformers import TFAutoModelForTokenClassification  # for tensorflow
from transformers import pipeline

model_name_or_path = "HooshvareLab/albert-fa-zwnj-base-v2-ner"  # Albert
tokenizer = AutoTokenizer.from_pretrained(model_name_or_path)
model = AutoModelForTokenClassification.from_pretrained(model_name_or_path)  # Pytorch
# model = TFAutoModelForTokenClassification.from_pretrained(model_name_or_path)  # Tensorflow

nlp = pipeline("ner", model=model, tokenizer=tokenizer)
example = "در سال ۲۰۱۳ درگذشت و آندرتیکر و کین برای او مراسم یادبود گرفتند."

ner_results = nlp(example)
print(ner_results)

📚 Documentation

Dataset Information

Property	Details
Model Type	Albert fine-tuned for NER
Training Data	Mixed dataset from ARMAN, PEYMA, and WikiANN

	Records	B-DAT	B-EVE	B-FAC	B-LOC	B-MON	B-ORG	B-PCT	B-PER	B-PRO	B-TIM	I-DAT	I-EVE	I-FAC	I-LOC	I-MON	I-ORG	I-PCT	I-PER	I-PRO	I-TIM
Train	29133	1423	1487	1400	13919	417	15926	355	12347	1855	150	1947	5018	2421	4118	1059	19579	573	7699	1914	332
Valid	5142	267	253	250	2362	100	2651	64	2173	317	19	373	799	387	717	270	3260	101	1382	303	35
Test	6049	407	256	248	2886	98	3216	94	2646	318	43	568	888	408	858	263	3967	141	1707	296	78

Evaluation

The following tables summarize the scores obtained by the model overall and per each class.

Overall

Model	Accuracy	Precision	Recall	F1
Albert	0.993405	0.938907	0.943966	0.941429

Per entities

Entity	Number	Precision	Recall	F1
DAT	407	0.820639	0.820639	0.820639
EVE	256	0.936803	0.984375	0.960000
FAC	248	0.925373	1.000000	0.961240
LOC	2884	0.960818	0.960818	0.960818
MON	98	0.913978	0.867347	0.890052
ORG	3216	0.920892	0.937500	0.929122
PCT	94	0.946809	0.946809	0.946809
PER	2644	0.960000	0.944024	0.951945
PRO	318	0.942943	0.987421	0.964670
TIM	43	0.780488	0.744186	0.761905

❓ Questions

Post a Github issue on the ParsNER Issues repo.

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