181 أسطر
3.4 KiB
Plaintext
181 أسطر
3.4 KiB
Plaintext
# Awesome Arabic NLP
|
|
|
|
> قائمة منسقة لأفضل **الأدوات والمكتبات والنماذج ومجموعات البيانات** في مجال
|
|
> **معالجة اللغة العربية (Arabic Natural Language Processing)**.
|
|
|
|
هذا المستودع يهدف إلى جمع أهم الموارد الخاصة بـ **Arabic NLP** في مكان واحد، وهو **مشروع مجتمعي** يمكن لأي شخص المساهمة فيه عبر Pull Request.
|
|
|
|
---
|
|
|
|
## Contents
|
|
|
|
* Frameworks & Libraries
|
|
* Named Entity Recognition (NER)
|
|
* Part-of-Speech Tagging (POS)
|
|
* Datasets
|
|
* Pre-trained Models
|
|
* Research Papers
|
|
* Learning Resources
|
|
* Contributing
|
|
|
|
---
|
|
|
|
# Frameworks & Libraries
|
|
|
|
### Python
|
|
|
|
* **CAMeL Tools**
|
|
https://github.com/CAMeL-Lab/camel_tools
|
|
Toolkit متقدم لمعالجة اللغة العربية (tokenization, morphology, NER, POS)
|
|
|
|
* **PyArabic**
|
|
https://github.com/linuxscout/pyarabic
|
|
مكتبة للتعامل مع النصوص العربية
|
|
|
|
* **AraNLP**
|
|
https://github.com/linuxscout/aranlp
|
|
أدوات متعددة لمعالجة اللغة العربية
|
|
|
|
* **Tashaphyne**
|
|
https://github.com/linuxscout/tashaphyne
|
|
مكتبة للتجذير والتحليل الصرفي
|
|
|
|
---
|
|
|
|
# Named Entity Recognition (NER)
|
|
|
|
التعرف على الكيانات الاسمية مثل:
|
|
|
|
* Person
|
|
* Location
|
|
* Organization
|
|
* Date
|
|
|
|
### Tools
|
|
|
|
* **CAMeL Tools NER**
|
|
https://github.com/CAMeL-Lab/camel_tools
|
|
|
|
* **Stanford Arabic NER**
|
|
https://stanfordnlp.github.io/CoreNLP/
|
|
|
|
### Datasets
|
|
|
|
* **WikiANN Arabic**
|
|
https://huggingface.co/datasets/wikiann
|
|
|
|
* **ARB-NER Dataset**
|
|
https://alt.qcri.org/resources/arb-ner/
|
|
|
|
---
|
|
|
|
# Part-of-Speech Tagging (POS)
|
|
|
|
تحديد نوع الكلمة داخل الجملة.
|
|
|
|
Examples:
|
|
|
|
* NOUN
|
|
* VERB
|
|
* ADJ
|
|
* ADV
|
|
* PRON
|
|
|
|
### Tools
|
|
|
|
* **CAMeL Tools POS Tagger**
|
|
https://github.com/CAMeL-Lab/camel_tools
|
|
|
|
* **Farasa POS Tagger**
|
|
https://farasa.qcri.org/
|
|
|
|
* **MADAMIRA**
|
|
https://github.com/ColumbiaNLP/madamira
|
|
|
|
* **Stanford POS Tagger**
|
|
https://stanfordnlp.github.io/CoreNLP/
|
|
|
|
### Datasets
|
|
|
|
* **UD Arabic Treebank**
|
|
https://universaldependencies.org/
|
|
|
|
---
|
|
|
|
# Datasets
|
|
|
|
* **SANAD Dataset**
|
|
https://data.mendeley.com/datasets/57zpx667y9
|
|
|
|
* **Arabic Poetry Dataset**
|
|
https://github.com/linuxscout/arabicpoetry
|
|
|
|
* **ArSAS Sentiment Dataset**
|
|
https://homepages.inf.ed.ac.uk/wmagdy/ArSAS.htm
|
|
|
|
* **Arabic SQuAD**
|
|
https://github.com/ppaudel/arabic-squad
|
|
|
|
---
|
|
|
|
# Pre-trained Models
|
|
|
|
* **AraBERT**
|
|
https://huggingface.co/aubmindlab/bert-base-arabert
|
|
|
|
* **AraGPT2**
|
|
https://huggingface.co/aubmindlab/aragpt2-base
|
|
|
|
* **AraELECTRA**
|
|
https://huggingface.co/aubmindlab/araelectra-base
|
|
|
|
* **CAMeL BERT**
|
|
https://huggingface.co/CAMeL-Lab
|
|
|
|
---
|
|
|
|
# Research Papers
|
|
|
|
* AraBERT: Transformer-based Model for Arabic NLP
|
|
https://arxiv.org/abs/2003.00104
|
|
|
|
* CAMeL Tools: An Open Source Toolkit for Arabic NLP
|
|
https://aclanthology.org/2020.lrec-1.868
|
|
|
|
* Farasa: A Fast and Accurate Arabic NLP Toolkit
|
|
https://aclanthology.org/L16-1170
|
|
|
|
---
|
|
|
|
# Learning Resources
|
|
|
|
* Natural Language Processing for Arabic (Book)
|
|
|
|
* Arabic Computational Linguistics
|
|
|
|
* NLP with Python
|
|
|
|
---
|
|
|
|
# Contributing
|
|
|
|
المشروع مفتوح للمجتمع ونرحب بإضافة أدوات أو مصادر جديدة.
|
|
|
|
### Steps
|
|
|
|
1. Fork the repository
|
|
2. Add your resource
|
|
3. Follow the existing format
|
|
4. Submit a Pull Request
|
|
|
|
Example:
|
|
|
|
```
|
|
- **Tool Name**
|
|
https://github.com/example/project
|
|
Short description
|
|
```
|
|
|
|
---
|
|
|