---

🎬 Movie Recommendation System

This project implements a movie recommendation system using unsupervised learning (K-Means Clustering) in a Jupyter Notebook environment. It clusters movies based on genres, ratings, and popularity, and then recommends similar movies within the same cluster.

---

📂 Project Structure

• movies_metadata.csv - Contains movie metadata including title, genres, rating, etc.
• movie_recommendation.ipynb - Jupyter Notebook implementing the entire pipeline from data cleaning to recommendation.

---

🛠️ Installation

Install the required Python libraries using:

pip install pandas numpy scikit-learn seaborn matplotlib

---

📥 Data Loading & Preprocessing

• Load movies_metadata.csv

• Extract relevant features:

  • title, genres, overview, poster_path, release_date, vote_average, popularity

• Drop rows with missing essential data

• Construct full poster image URLs using TMDB's base URL (https://image.tmdb.org/t/p/w500)

---

🧠 Feature Engineering

• Genre Parsing: Convert genre strings into Python lists of genre names
• One-Hot Encoding: Encode genres with MultiLabelBinarizer
• Normalization: Scale vote_average and popularity with MinMaxScaler
• Final Features: Concatenate genre and numeric features into one DataFrame for clustering

---

📊 Clustering

• Apply KMeans clustering (k = 2 to 20) to find optimal clusters

• Evaluate using:

  • Inertia (Elbow Method)
  • Silhouette Score

• Visualize both to choose the best number of clusters

---

🧼 Cluster Analysis

• Apply final KMeans model (default: k=10)
• Assign each movie to a cluster
• Visualize movie count per cluster to assess distribution

---

📈 Genre Distribution

• Explode the genre lists and count each genre’s frequency
• Visualize as a bar chart using Seaborn

---

🤖 Movie Recommendation Function

A simple content-based recommendation using clustering:

Testing:

recommend_movies("Iron Man", n=5)

Returns:

• Title
• Overview
• Release Date
• Vote Average
• Popularity
• Poster Image URL

Example output:

🎬 Say Nothing (2001-08-01)
⭐ Rating: 9.0 | 🔥 Popularity: 0.22
📝 Housewife Grace Needham has a one-night stand...
🖼️ Poster: https://image.tmdb.org/t/p/w500/e5zFhKKniiiep8oM7Y3g6gLygHS.jpg

---

📌 Notes

• All code is contained in a single Jupyter Notebook.
• Ensure that movies_metadata.csv is in the same directory.
• Ideal for content-based recommendation systems where genre similarity is key.

---

📷 Screenshots (Optional)

Include charts like:

• Elbow Plot
• Silhouette Score Plot
• Cluster Distribution
• Genre Frequency Plot

---

📜 License

MIT License — free to use and modify with credit.

---