A beginner-level content-based recommendation system using the Netflix dataset. This project covers basic data cleaning, similarity calculations, and a simple recommender engine. Although the system wasn’t completed due to increasing complexity, it serves as a valuable introduction to recommendation systems for beginners.
🎬 Netflix Content-Based Recommender 📌 Objective: Recommend similar movies or TV shows based on metadata like description, cast, director, and genre.
📁 Dataset Source: Netflix Titles Dataset on Kaggle
⚙️ Features Used: description (cleaned with NLP)
cast, director, and genre (combined into one metadata column)
Cosine Similarity used for content similarity
🚀 How to Use: Run the notebook and use:
get_recommendations("Narcos")
To get 10 most similar titles.
🙋♂️ Author
Umeir Mohamed
Master’s Student in Data Science – Milano Bicocca University
LinkedIn | GitHub