🎬 Content-Based Movie Recommendation System

This is a content-based movie recommender system built using NLP techniques and cosine similarity to recommend movies based on user-selected titles. The app is deployed using Streamlit and fetches real-time movie posters from the OMDb API. For similarity.pkl file you have to run code where we dump file. Because file size is more than 100 MB. While Github allow only 25MB max size.

Kaggle Dataset:- https://www.kaggle.com/datasets/rounakbanik/the-movies-dataset

🚀 Features

• 📌 Clean NLP Pipeline (Tokenization + Stemming)
• 🧠 Content-Based Filtering (Using Genres, Cast, Director, Keywords, Overview)
• 🔍 Cosine Similarity with CountVectorizer
• 🎥 Poster Fetch via OMDb API
• 🖥️ Streamlit Web Interface
• 💾 Uses Pickle for Fast Inference

🛠️ Tech Stack

📁 Folder Structure

📁 data

📷 Screenshots

🔹 App Homepage

🔹 Top 5 Recommendations

💡 How It Works

1. Load IMDB dataset and clean it. https://www.kaggle.com/datasets/rounakbanik/the-movies-dataset
2. Extract features: overview, genres, keywords, cast, crew.
3. Convert text into vectors using CountVectorizer.
4. Compute cosine similarity between vectors.
5. Recommend top 5 most similar movies to the selected one.
6. Use OMDb API to fetch movie posters by IMDb ID.

📦 Installation & Run Locally

1. Clone the repo