📖 PDF Question-Answering App (RAG-based)

A PDF Question-Answering App built with RAG (Retrieval-Augmented Generation), allowing users to upload PDFs and ask context-based questions.
Powered by Streamlit, LangChain, Ollama, and Chroma/FAISS for efficient and accurate answers.

---

📌 Project Overview

This project is a Retrieval-Augmented Generation (RAG) based Question-Answering System where users can upload a PDF document and ask questions related to its content.

It extracts the text, splits it into chunks, stores it in a vector database, and uses a Large Language Model (LLM) to provide accurate, context-aware answers.
The RAG pipeline ensures that answers are based on real document context to minimize hallucination.

---

✨ Key Features

• 📂 Upload PDF and extract text automatically.
• 🔍 Context-aware Q/A using RAG (retrieves relevant chunks before answering).
• ⚡ Efficient Vector Search with Chroma/FAISS backend.
• 🤖 LLM-powered answers using Ollama + Llama 3.1.
• 🖥️ Interactive GUI built with Streamlit for seamless use.
• 🔒 Handles temporary files securely during processing.

---

🛠️ Tools & Technologies

• – Interactive web interface
• – RAG pipeline
• – Local LLM (llama3.1, nomic-embed-text)
• – Vector similarity search
• – Extract text from PDFs

---

📋 Prerequisites

Before running the project, make sure you have:

• 🐍 Python 3.9+ installed
• 📦 Ollama installed and running locally → Install Ollama
• 📥 Pull required models in Ollama:

  ollama pull llama3.1
  ollama pull nomic-embed-text

---

📂 Project Structure

The repository is organized as follows:

pdf-qa-rag/
│── app.py              
└── requirements.txt    

---

🚀 Quick Start

Follow these steps to set up and run the project locally:

1️⃣ Clone the Repository

git clone https://github.com/muqadasejaz/pdf-qa-rag-system.git

2️⃣ Create a Virtual Environment (Recommended)

# For Windows
python -m venv venv
venv\Scripts\activate

3️⃣ Install Dependencies

pip install -r requirements.txt

4️⃣ Install Ollama & Pull Required Models

• Download and install Ollama

• Pull models locally:

ollama pull llama3.1
ollama pull nomic-embed-text

5️⃣ Run the Application

streamlit run app.py

6️⃣ Open in Browser

• Streamlit will provide a local URL (e.g. http://localhost:8501).
• Open it in your browser to access the PDF Question-Answering App.

---

📱 How to Use

1. Start the app by running:

   streamlit run app.py

2. Open the Streamlit link in your browser (default: http://localhost:8501).

3. 📂 Upload a PDF

• Use the file uploader to select any PDF (e.g., notes, research papers, or reports).

4. ⚡ Processing

• The system will extract text, split it into chunks, and store them in the vector database.

5. 📝 Ask Your Question

• Enter your query in the text box (e.g., "What is the main conclusion of this paper?").

6. 💡 Get the Answer

• The LLM (Ollama + LangChain) retrieves the most relevant context and generates an accurate, context-based answer.

---

🏗️ Architecture

The system follows a RAG (Retrieval-Augmented Generation) pipeline for answering questions from PDFs.

Workflow:

1. 📂 PDF Documents → Uploaded by the user.
2. 🔎 LangChain Pipeline → Handles:
   • PDF Text Extraction
   • Text Chunking
   • Vector Store (Chroma/FAISS)
3. 🤖 Ollama LLM → Uses llama3.1 for generating answers.
4. 💡 Generated Response → Delivered back to the user.

---

🖥️ GUI Preview

The application comes with a simple and interactive Streamlit-based UI.

📌 Home Screen

Upload your PDF, enter your question, and get instant answers.

📌 PDF Upload & Processing

After uploading, the system extracts and processes the PDF text for querying.

📌 Q&A Result

Ask questions, and the system provides precise answers from the document.

• Answer Query 1:

• Answer Query 2:

• Output:

• Wrong Query:

  

---

🙏 Acknowledgements

This project is built with the help of:

• LangChain – for chaining components in the RAG pipeline
• Ollama – for running LLaMA 3.1 locally
• Streamlit – for the user-friendly interface
• ChromaDB / FAISS – as vector databases for semantic search
• nomic-embed-text – for text embeddings

Special thanks to the open-source community for providing tools that make projects like this possible 🚀

---

👤 Author

Muqadas Ejaz

BS Computer Science (AI Specialization)

AI/ML Engineer

Data Science & Gen AI Enthusiast

📫 Connect with me on LinkedIn

🌐 GitHub: github.com/muqadasejaz

---

📎 License

This project is open-source and available under the MIT License.