Chiraag-P-V/bank_customer_ticket_category_classifier_fine_tuned

Bank Customer Ticket Category Classifier (Fine-Tuned DistilBERT)

Model Description

This model is a fine-tuned version of distilbert-base-uncased for bank customer complaint classification.
It classifies complaints into one of three categories:

1. Credit card or prepaid card
2. Checking or savings account
3. Mortgage

The model was developed to help banks and financial institutions automatically tag and route complaints to the correct department, improving efficiency, accuracy, and response times.

---

Intended Uses & Limitations

Intended Use Cases

• Automating complaint classification in customer service systems.
• Categorizing historical complaint datasets for analytics.
• Integrating with chatbots or CRM systems for real-time tagging.

Limitations

• Only supports English-language inputs.
• Designed specifically for the three categories above — other categories will not be classified correctly.
• May underperform on slang-heavy or incomplete sentences.

---

Example Inference

from transformers import pipeline

classifier = pipeline("text-classification", model="Chiraag-P-V/bank_customer_ticket_category_classifier_fine_tuned")

text = "I have a credit card issue."
result = classifier(text)

print(result)
# Example output:
# [{'label': 'Credit card or prepaid card', 'score': 0.987}]