LLM Bill Chat App

This project is a proof of concept for a chat application utilizing a Large Language Model (LLM) to assist users with their telecom billing inquiries. The application is built using Python and Streamlit, providing an interactive web interface for users to engage with.

Features

Maintain chat conversation context (history)
Allow users to query their billing information
Compare the last bills and provide insights
Respond exclusively with the user's own billing information
Augment the prompt instructions with user's text recognized entities - NER -> Dynamic prompting# Aplicația LLM Bill Chat

Project Structure

hf-llm-bill-chat/ │ ├── app.py ├── pages/ │ ├── One_model.py │ ├── Two_models.py │ ├── data/ │ └── user_data/ │ └── user_data_724077190.json └── logs/ └── conversation_logs.json

Code Implementation - Main Script

Initialize Streamlit app
Load user data and display existing bills
Handle file uploads for new bills
Maintain chat conversation context
Leverage user text entities recognition (NER) for dynamic prompting
Call OpenAI models (GPT-4o and GPT-4o-mini) for responses
Display responses in two parallel columns
Log conversation details

Future Improvements

Enhance Natural Language Processing (NLP) capabilities for better entity recognition
Implement user authentication for secure access to billing information
Add support for more telecom providers and bill formats
Improve the user interface for better user experience
Integrate with payment gateways for bill payments

Conclusion

The LLM Bill Chat App demonstrates the potential of using Large Language Models for telecom billing assistance.
The app provides a user-friendly interface for querying billing information and getting insights.
Future improvements can enhance the app's capabilities and user experience.