🏷️ arabic-ner-v2

arabic-ner-v2 is an enterprise-grade Named Entity Recognition model for Modern Standard Arabic (MSA) and Egyptian dialect texts, detecting Persons, Organizations, Locations, and Dates.

Developed by Yousef Khaled Gouda.

📊 Evaluation Metrics

• Overall Macro F1: 0.928
• PER F1: 0.941 | ORG F1: 0.922 | LOC F1: 0.935 | DATE F1: 0.914

💻 Usage Example

from transformers import pipeline

ner = pipeline("ner", model="Yousef1258852/arabic-ner-v2", aggregation_strategy="simple")
text = "التقى يوسف خالد بممثلي شركة أرامكو في القاهرة يوم الأحد الماضي."
print(ner(text))