ikramknd/dziribert-ecommerce-intent-detection

DziriBERT E-commerce Intent Detection

This model is a fine-tuned version of alger-ia/dziribert for multi-label intent detection in e-commerce customer service conversations.

Model Description

• Base Model: DziriBERT
• Task: Multi-label Intent Classification
• Number of Intents: 28
• Language: Arabic (including Arabizi)

Intents Covered

• greeting
• goodbye
• store_info
• products_catalogue_request
• emojis
• price
• stock
• order_taking
• confirmation
• modification
• cancellation
• delivery_question
• order_tracking
• feedback
• human_assistance
• ask_for_description
• produit_info_description
• produit_info_color
• produit_info_size
• produit_info_brand
• produit_info_promotion
• quantity
• store_existence
• store_location
• store_hours
• other_store_info
• feedback_positive
• feedback_negative

Usage

from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch

model_name = "ikramknd/dziribert-ecommerce-intent-detection"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSequenceClassification.from_pretrained(model_name)

# Example
text = "مرحبًا، كم سعر هذا المنتج؟"
inputs = tokenizer(text, return_tensors="pt", truncation=True, padding=True)
outputs = model(**inputs)
probs = torch.sigmoid(outputs.logits)