|
--- |
|
license: mit |
|
language: |
|
- fa |
|
--- |
|
--- |
|
license: mit |
|
--- 📚 Persian Multi-Source Dataset Collection |
|
🌟 Dataset Overview |
|
A comprehensive Persian language dataset combining diverse sources to create a rich training corpus for language models. |
|
|
|
📊 Data Sources |
|
🌐 Public Datasets |
|
MaralGPT Collections |
|
Persian blogs |
|
Persian quotes |
|
Story Collections |
|
TinyStories-Farsi |
|
FarsiTinyStories |
|
News & Media |
|
Farsi news corpus |
|
Persian daily news |
|
Persian blog posts |
|
Conversation & QA |
|
Persian QA pairs |
|
Telegram channel content |
|
Specialized Content |
|
ZharfaTech Open Platypus |
|
ZharfaTech OpenAssistant Guanaco |
|
Paraphrase detection corpus |
|
ParsinLu entailment dataset |
|
💬 Custom Collections |
|
Nini Site Data |
|
User comments |
|
Discussion threads |
|
User issues and queries |
|
📈 Statistics |
|
Multiple genres and writing styles |
|
Diverse vocabulary range |
|
Rich conversational patterns |
|
Real-world language usage |
|
🎯 Use Cases |
|
Language Model Training |
|
Conversational AI |
|
Text Analysis |
|
Persian NLP Research |
|
✨ Features |
|
🔄 Merged and cleaned data |
|
📝 Natural language variations |
|
💡 Rich contextual information |
|
🗣️ Authentic Persian conversations |
|
Perfect for training Persian language models with diverse, real-world language patterns! |
|
|
|
#PersianNLP #Dataset #MachineLearning #NaturalLanguageProcessing #DataScience |
|
|
|
|
