🤖 Amazon RAG Chatbot (LangChain + Gemini + FAISS)

📌 Project Overview

This project is a Retrieval-Augmented Generation (RAG) Chatbot built using LangChain, Google Gemini API, and FAISS Vector Database.

The chatbot answers user queries only based on provided data, ensuring accurate and context-based responses instead of general AI knowledge.

---

🚀 Features

• 🔍 RAG Architecture (Retrieval + Generation)
• 🧠 Uses Google Gemini API for intelligent responses
• 📚 FAISS Vector Database for fast similarity search
• 📄 Reads data from external file (docs.txt)
• 💬 Interactive Streamlit Chat UI
• 🔌 Flask API for backend communication
• ❌ Prevents hallucination (answers only from data)

---

🏗️ Project Structure

Amazon_Chat_Bot/
│
├── rag_core.py        # RAG logic (retrieval + generation)
├── api.py             # Flask backend API
├── app.py             # Streamlit frontend UI
├── data/
│   └── docs.txt       # Knowledge base
├── requirements.txt
└── README.md

---

⚙️ Installation

1️⃣ Clone Repository

git clone <your-repo-url>
cd Amazon_Chat_Bot

2️⃣ Install Dependencies

pip install -r requirements.txt

---

🔑 Setup API Key

Get your Gemini API key from Google AI Studio and set it in your code:

os.environ["GOOGLE_API_KEY"] = "YOUR_API_KEY"

---

▶️ Run the Project

Step 1: Start Flask API

python api.py

Step 2: Start Streamlit UI

python -m streamlit run app.py

Step 3: Open in Browser

http://localhost:8501

---

📄 Knowledge Base

The chatbot reads data from:

data/docs.txt

Example content:

• Amazon return policy
• Order tracking steps
• Account login instructions
• Prime membership details
• Customer support info

---

🧠 How It Works

1. User enters a query
2. Query is converted into embeddings
3. FAISS retrieves relevant documents
4. Gemini API generates answer using context
5. Response is returned to UI

---

📊 Technologies Used

• Python 🐍
• LangChain
• Google Gemini API
• FAISS
• Streamlit
• Flask

---

🎯 Use Cases

• Customer Support Chatbot
• FAQ Automation
• E-commerce Assistant
• Knowledge Base Search

---

⚠️ Limitations

• Depends on quality of input data
• Requires API key
• Limited by small dataset (can be expanded)

---

🚀 Future Improvements

• 📄 PDF Upload Support
• 🎤 Voice Assistant
• 🌐 Deployment (Render / AWS)
• 🎨 Advanced UI (ChatGPT-style)
• 🔐 User Authentication

---

👨‍💻 Author

Kalavanthula Sekhar Sanskriti University

---

📜 License

This project is for educational purposes.