🎬 "Netflix EDA with Content-Based Recommendations”

Netflix Data Analysis & Recommendation System with Python 📊🔍

Dive into the world of Netflix with Python! Analyze over 10,000+ titles and uncover trends, visual stories, and content similarities from your favorite streaming giant using data science and machine learning.

---

📌 Overview

Netflix, one of the world’s leading video streaming platforms, has a massive global library of movies 🎥 and TV shows 📺, serving 222M+ subscribers (as of mid-2021). This project delivers a complete exploratory data analysis (EDA) of Netflix content and a content-based recommendation engine using Python.

You'll discover insights about genres, ratings, countries, time trends, and even get recommendations for what to watch next — all using real-world Netflix data.

---

🔍 Project Highlights

🧰 Libraries Used

• Pandas 🐼 – data manipulation
• NumPy 🧮 – numerical operations
• Matplotlib 📊 – visualizations
• Seaborn 📈 – statistical plotting
• WordCloud ☁️ – generate textual data clouds
• Scikit-learn 🤖 – ML & similarity computation

---

📊 Dataset Columns

Column	Description
show_id	Unique identifier for each title
type	Movie or TV Show
title	Title of the content
director	Director’s name
cast	Main cast members
country	Country of production
date_added	Date added to Netflix
release_year	Year of original release
rating	Maturity rating (e.g., TV-MA, PG)
duration	Duration in minutes or number of seasons
listed_in	Genre(s) of the content
description	Short summary or synopsis

---

📈 Key Insights & Visualizations

• 🌍 Country-wise Rating Distribution
• 🎭 Genre Trends by Country
• 🔞 Genre vs. Rating Matrix
• 📊 Correlation Heatmaps
• 🎤 Most Active Actors
• ⏱️ Content Duration Analysis
• 📅 Time Trends in Content Addition
• 👶 Age Group Classification
• 🎬 Top Genres & Directors
• 🌎 Global Content Distribution
• 🍿 Movie vs. TV Show Breakdown

---

🤖 NEW: Content-Based Recommendation Engine

A lightweight machine learning engine recommends similar titles based on the title, genre, and description of Netflix content.

recommend("Stranger Things", n=5)

🎯 Suggests similar shows by comparing textual similarity using TF-IDF and cosine similarity. Great for demo, search enhancement, and personalized recommendations.

---

🎯 Goals of the Project

• Deliver intuitive, visual summaries of Netflix’s library
• Practice real-world EDA and text-based ML techniques
• Analyze how genres, countries, and ratings shape content strategy
• Explore growth and evolution of Netflix over the years
• Introduce a content recommender for viewers and analysts

---

🌟 Features

✅ Cleaned and preprocessed dataset ✅ Beautiful plots with Seaborn and Matplotlib ✅ Word clouds for cast, genres, and descriptions ✅ Year-wise and country-wise analysis ✅ TF-IDF–based recommendation system ✅ Easy-to-use recommend("Title") function

---

🔮 Future Enhancements

• 📊 Add interactive dashboards (Streamlit / Plotly Dash)
• 🎥 Sentiment analysis on content descriptions
• 🧠 Expand to collaborative filtering
• 🔄 Integrate with Netflix API or scrape for live updates
• 🔁 Compare with Prime Video, Hulu, Disney+

---

🤝 Contributors

• Esha Yalagi
• Mudabbir

---

🧠 Topics & Skills

#ExploratoryDataAnalysis #DataVisualization #MachineLearning #ContentRecommendation #Netflix #Python #Pandas #Seaborn #ScikitLearn #WordCloud

---

📣 About

Explore Netflix’s World 🌍🍿 An engaging, visual deep-dive into Netflix’s content library using Python. Whether you're a data enthusiast, student, or Netflix binge-watcher, this project helps you uncover the data science behind content trends and intelligent recommendations.

---