README.md

metadata

title: Persian NER
colorFrom: green
colorTo: blue
sdk: gradio
sdk_version: 4.44.0
app_file: app.py
pinned: false

Persian Named Entity Recognition (NER)

This space performs Named Entity Recognition on Persian (Farsi) text using the ParsBERT model.

Features

• Identifies person names, organizations, locations, dates, times, money, and percentages
• Highlights entities with color-coding
• Runs on CPU (no GPU required)
• Provides confidence scores for each detected entity

Model

• Model: HooshvareLab/bert-base-parsbert-ner-uncased
• Framework: Transformers + PyTorch
• Language: Persian (Farsi)

Usage

Simply enter Persian text in the input box and click "Analyze Text" to see the detected entities highlighted in the text.

Entity Types

• PER (شخص): Person names
• ORG (سازمان): Organizations
• LOC (مکان): Locations
• DAT (تاریخ): Dates
• TIM (زمان): Times
• MON (پول): Money amounts
• PCT (درصد): Percentages

Citation

@article{ParsBERT,
  title={ParsBERT: Transformer-based Model for Persian Language Understanding},
  author={Mehrdad Farahani, Mohammad Gharachorloo, Marzieh Farahani, Mohammad Manthouri},
  journal={ArXiv},
  year={2020},
  volume={abs/2005.12515}
}