🎬 Movie Recommendation System

A Machine Learning powered Content-Based Movie Recommendation System built using Streamlit, Scikit-Learn and TMDb API.

📌 Overview

The Movie Recommendation System is an interactive Machine Learning web application that recommends movies similar to a user's selected movie using Content-Based Filtering.

The recommendation engine analyzes movie genres using TF-IDF Vectorization and computes similarity scores using Cosine Similarity to identify movies with similar content.

To provide a richer user experience, the application integrates the TMDb API to fetch:

🎬 Movie Posters
⭐ TMDb Ratings
📅 Release Dates
📖 Movie Overview
🌍 Original Language
🔥 Popularity Score
🗳 Vote Count

The application is developed using Streamlit, providing a modern, responsive, and interactive web interface.

🚀 Features

🤖 Machine Learning

Content-Based Recommendation System
TF-IDF Vectorization
Cosine Similarity
Top-N Similar Movie Recommendations

🎥 Movie Information

Movie Posters
Movie Overview
Release Date
TMDb Rating
Original Language
Popularity
Vote Count
Average User Rating

🎨 User Interface

Modern Dark Theme
Interactive Streamlit Dashboard
Responsive Layout
Sidebar Statistics
Recommendation Cards
Similarity Progress Bar
Genre Badges
Movie Search
Professional UI Design

⚡ Performance

Cached TMDb API Calls
Session State Management
Error Handling
Missing Poster Fallback
Fast Recommendation Retrieval

📊 Dataset

This project uses the MovieLens Latest Small Dataset.

Dataset	Count
Movies	9,742
Users	610
Ratings	100,836
Tags	3,683+

Dataset includes:

Movies
Ratings
Tags
TMDb Links

🧠 Machine Learning Workflow

Movie Dataset
      │
      ▼
Data Preprocessing
      │
      ▼
Genre Extraction
      │
      ▼
TF-IDF Vectorization
      │
      ▼
Cosine Similarity Matrix
      │
      ▼
Top Similar Movies
      │
      ▼
TMDb API
      │
      ▼
Interactive Streamlit Dashboard

📖 Recommendation Algorithm

The recommendation pipeline consists of the following steps:

User selects a movie.
Movie genres are converted into TF-IDF vectors.
Cosine similarity is computed between movies.
Top similar movies are selected.
Movie metadata is fetched from TMDb.
Recommendations are displayed with posters and ratings.

🛠 Tech Stack

Category	Technology
Programming Language	Python
Machine Learning	Scikit-Learn
Data Processing	Pandas, NumPy
Web Framework	Streamlit
API	TMDb API
Model Serialization	Joblib
Visualization	Matplotlib

📂 Project Structure

Movie_Recommendation_System/

│

├── app/
│   └── app.py

├── dataset/
│   ├── movies.csv
│   ├── ratings.csv
│   ├── links.csv
│   └── tags.csv

├── model/
│   ├── cosine_similarity.pkl
│   ├── movie_indices.pkl
│   └── movies.pkl

├── notebooks/
│   └── movie_recommendation.ipynb

├── output/
│   ├── Home Page.png
│   ├── Selected Movie.png
│   ├── Recommendations.png
│   ├── average_rating_distribution.png
│   ├── genre_distribution.png
│   ├── rating_distribution.png
│   ├── ratings_per_movie.png
│   ├── ratings_per_user.png
│   ├── top_rated_movies.png
│   └── top_users.png

├── requirements.txt
├── README.md
└── .gitignore

📷 Application Screenshots

🏠 Home Page

🎬 Selected Movie

🍿 Movie Recommendations

📊 Exploratory Data Analysis

Average Rating Distribution

Genre Distribution

Rating Distribution

Ratings Per Movie

Ratings Per User

Top Rated Movies

Top Users

📈 Results

✔ Successfully recommends similar movies.

✔ Real-time movie posters using TMDb.

✔ Displays ratings, release dates and overview.

✔ Professional interactive dashboard.

✔ Supports more than 9,700 movies.

✔ Average response time below 2 seconds after caching.

💻 Installation

Clone the repository

git clone https://github.com/YOUR_USERNAME/Movie_Recommendation_System.git

Navigate to project folder

cd Movie_Recommendation_System

Create virtual environment

python -m venv .venv

Activate virtual environment

Windows

.venv\Scripts\activate

Linux/macOS

source .venv/bin/activate

Install dependencies

pip install -r requirements.txt

🔑 TMDb API Configuration

Create a folder named:

.streamlit

Create a file:

secrets.toml

Add your TMDb credentials:

TMDB_API_KEY="YOUR_API_KEY"

TMDB_BEARER_TOKEN="YOUR_BEARER_TOKEN"

▶ Running the Application

streamlit run app/app.py

🐞 Troubleshooting

Posters not loading

Verify your TMDb API credentials.
Check your internet connection.
Ensure tmdbId exists for the selected movie.

Missing Model Files

Verify that the following files exist:

movies.pkl
movie_indices.pkl
cosine_similarity.pkl

Slow Loading

The first TMDb request may take longer.

Subsequent requests are cached automatically.

🚀 Future Improvements

Hybrid Recommendation System
Collaborative Filtering
Deep Learning Recommendation Model
Personalized Recommendations
User Authentication
Favorite Movies
Watchlist
Movie Trailer Integration
Actor & Director Information
Genre & Year Filters
Deploy on Streamlit Community Cloud

📚 Learning Outcomes

Through this project, I gained practical experience in:

Machine Learning Recommendation Systems
TF-IDF Vectorization
Cosine Similarity
Content-Based Filtering
Data Preprocessing
API Integration
Streamlit Application Development
Session State Management
Model Serialization
Interactive Dashboard Design

👨‍💻 Author

Harsh Vardhan Chaudhary

B.Tech Computer Science & Engineering
SRM Institute of Science and Technology

🌐 Connect With Me

LinkedIn

https://www.linkedin.com/in/connect-harsh-vardhan

GitHub

https://github.com/harshvardhan4096

⭐ Support

If you found this project useful, consider giving this repository a ⭐ on GitHub.

It motivates me to build and share more Machine Learning and AI projects.

📄 License

This project is licensed under the MIT License.

Name		Name	Last commit message	Last commit date
Latest commit History 1 Commit
.streamlit		.streamlit
app		app
dataset		dataset
notebook		notebook
output		output
.gitignore		.gitignore
LICENSE		LICENSE
README.md		README.md
requirements.txt		requirements.txt

Folders and files

Latest commit

History

Repository files navigation