AI Powered Article QnA

Streamlit-Based URL Embedding and Q&A System

A modern web application built with Streamlit that allows users to paste URLs, extract embeddings from the content, and use a Retrieval-Augmented Generation (RAG) approach to answer user questions based on the articles' content.

Features

• URL Input: Paste up to 3 URLs to extract content.
• Embedding and RAG: Generate embeddings for the articles and use RAG for enhanced question answering.
• Hugging Face Integration: Leverage models from Hugging Face for embeddings and language modeling.
• Local Model Alternative: Option to use local models like Ollama Mistral.
• User-Friendly Interface: Simple and intuitive Streamlit-based UI.

Demo

Tech Stack

• Streamlit
• LangChain
• Hugging Face
• Ollama
• Mistral
• Python
• FAISS

Installation

1. Clone the Repository:

   git clone https://github.com/PrakharMishra531/AI-powered-article-QnA
   cd your-repo

2. Create and Activate Virtual Environment:

   python -m venv env
   source env/bin/activate   # On Windows use `env\Scripts\activate`

3. Install Dependencies:

   pip install -r requirements.txt

4. Set Up Environment Variables: Create a .env file and add your Hugging Face API token and other necessary configurations.

   HUGGINGFACE_API_KEY=your_api_key_here

Usage

1. Run the Streamlit App:

   streamlit run app.py

2. Interact with the Application:

   • Open the web browser at the provided local URL.
   • Paste up to 3 URLs.
   • Ask questions and get answers based on the content of the URLs.