This is a fine-tuned version of UBC-NLP/MARBERTv2, specifically adapted for intent classification in Egyptian Colloquial Arabic, with a primary focus on e-commerce and customer service scenarios.

It enables accurate understanding of user queries in dialectal Arabic, empowering applications like chatbots, support assistants, and ticket routing systems.

🧠 Model Details

Model Type: bert-for-sequence-classification
Base Model: UBC-NLP/MARBERTv2
Language: Arabic (Egyptian dialect)
Developer: Shaza Aly
License: Apache 2.0
Repository: https://huggingface.co/ShazaAly/syplyd-marbert-1

🚀 Usage

This model can be used directly with the Hugging Face transformers library:

python

from transformers import pipeline

classifier = pipeline("text-classification", model="ShazaAly/syplyd-marbert-1")

# Example 1
text_1 = "عايز أعرف الأوردر بتاعي هيوصل امتى؟"
print(classifier(text_1))
# Output: [{'label': 'track_order_status', 'score': ...}]

# Example 2
text_2 = "المنتج ده غالي، فيه بديل أرخص؟"
print(classifier(text_2))
# Output: [{'label': 'product_alternatives', 'score': ...}]