Movie Maven Chatbot

A production-grade CLI Chatbot that answers questions about 45,000+ movies using LangGraph, Google Gemini, and FAISS.

Features

• Structured Queries: Accurate answers for budget, revenue, release dates, and cast.
• Semantic Search: Find movies by plot, theme, or vague description (e.g., "movies about time travel").
• Multi-turn Conversations: Maintains context (e.g., "Who directed it?").
• Local Vectors: Uses FAISS and HuggingFace embeddings (efficient, no extra API costs).

Prerequisites

• Dataset: movies_metadata.csv, keywords.csv, credits.csv in the dataset/ directory.
• Google API Key: Needed for the Chatbot.

---

Option 1: Run with Docker (Recommended)

This method handles all dependencies and data processing automatically inside a container.

1. Setup Environment Create a .env file:

   GOOGLE_API_KEY=your_api_key_here

2. Run the Bot

   docker-compose run --rm movie-bot

   The container will automatically check for processed data. If missing, it will process the CSVs (takes ~2-3 mins) before starting the chat.

3. Interact The chatbot will start in your terminal. Type quit to exit.

---

Option 2: Run Locally (No Docker)

1. Install Python 3.11+

2. Install Dependencies

   pip install -r requirements.txt

3. Setup Environment Create a .env file:

   GOOGLE_API_KEY=your_api_key_here

4. Process Data (One-time setup)

   python src/data_processor.py

5. Run the App

   python main.py

Architecture

• LLM: Google Gemini 1.5 Flash.
• Agent: LangGraph StateGraph.
• Vector Store: FAISS with all-MiniLM-L6-v2 embeddings.