🎬 Movie Recommendation System

This repository contains a Movie Recommendation System built with Python, leveraging machine learning techniques to suggest movies based on user preferences. The model employs cosine similarity on a matrix of movie ratings, enabling recommendations for similar movies based on a selected title.

📂 Project Overview

The system uses the MovieLens 20M Dataset for training, which includes millions of movie ratings. Our approach focuses on generating high-quality recommendations by creating a movie-user matrix and applying a collaborative filtering technique.

✨ Key Features

• Cosine Similarity Calculation: Computes similarity scores between movies based on user ratings.
• Genre One-Hot Encoding: Extracts and utilizes genre data for better filtering.
• Easy-to-Use Function: Just input a movie name, and get the top recommendations instantly.
• Data Preprocessing: Handles missing values, duplicates, and applies filtering to retain only movies with a significant number of ratings.

🔧 Installation & Setup

1. Clone the Repository:

   git clone https://github.com/yourusername/movierecommendation.git
   cd movierecommendation

2. Install Dependencies:
   Make sure you have Python installed, then install the required packages.

   pip install -r requirements.txt

3. Download Dataset:
   The dataset is automatically downloaded via Kaggle API when running the script. Ensure you have Kaggle API configured.

🛠️ Usage

• Load and Run: After installation, you can run the script directly to test the recommendation function.

  # Import the main function
  from movie_recommendation import get_recommendations_by_name
  
  # Get recommendations
  recommendations = get_recommendations_by_name("Thor", similarity_matrix, movie_rating, top_n=5)
  print(recommendations)

📊 Methodology

• Data Preprocessing: Cleans the dataset, handling missing values and duplicates.
• Matrix Generation: Creates a sparse matrix using csr_matrix to optimize memory.
• Similarity Calculation: Applies cosine similarity to find the closest matches for each movie in the dataset.

👤 Author

• Prabhat Ghimire - LinkedIn | GitHub

📜 License

This project is licensed under the MIT License - see the LICENSE file for details.

💡 Contributing

Feel free to submit issues, fork the repository, and make a pull request with your ideas for improvement.