Choosing the right movie to watch can be frustrating. Recommendation systems, like those used by **Netflix and Prime Video**, help by suggesting movies based on user interests.

This blog shares how I built a **Movie Recommendation System** using **Python, Streamlit, and the TMDb API**. The goal was to create a simple yet effective system that:

✅ **Suggests movies similar to the one selected**

✅ **Fetches movie posters dynamically**

✅ **Has a clean and interactive UI**

✅ **Can be deployed online for free**

---

## **Technologies Used**

For this project, I used:

🔹 **Python** – The core programming language

🔹 **Pandas & NumPy** – For data handling and processing

🔹 **Scikit-Learn** – To build the recommendation algorithm

🔹 **TMDb API** – To fetch movie posters dynamically

🔹 **Streamlit** – To build the interactive web app

🔹 **Git & GitHub** – To manage and deploy the project

🔹 **Git LFS** – To handle large datasets

---

## **Step 1: How the Recommendation System Works**

A **Content-Based Filtering** approach was implemented to recommend movies based on their similarity to a selected movie.

### **Process:**

1️⃣ **User selects a movie** from the dropdown.

2️⃣ **The system finds similar movies** using machine learning techniques.

3️⃣ **Movie posters are fetched** using the TMDb API.

4️⃣ **Results are displayed** in an interactive UI.

This ensures relevant recommendations while keeping the experience visually engaging.

---

## **Step 2: Fetching Movie Posters with TMDb API**

To enhance the UI, the **TMDb API** was used to fetch movie posters.

```python

import requests

API_KEY = "your_tmdb_api_key"

def get_movie_poster(movie_id):

url = f"https://api.themoviedb.org/3/movie/{movie_id}?api_key={API_KEY}"

response = requests.get(url).json()

poster_path = response["poster_path"]

full_poster_url = f"https://image.tmdb.org/t/p/w500{poster_path}"

return full_poster_url

This function takes a movie ID, calls the TMDb API, and returns the poster URL, which is then displayed in the Streamlit app.

---

Step 3: Deploying the App for Free

The app was deployed on Streamlit Community Cloud, a free hosting platform for Streamlit apps.

Steps to Deploy:

1️⃣ Push the Code to GitHub

git init

git add .

git commit -m "First commit"

git branch -M main

git remote add origin https://github.com/Sourabh-Kumar04/Raso-Movie-Recommendation.git

git push -u origin main

2️⃣ Deploy on Streamlit Cloud

Go to Streamlit Community Cloud

Click on "New app"

Select the GitHub repository

Click "Deploy"

The app went live within minutes. 🎉

🔗 Live App: Raso Movie Recommendation

---

Step 4: Handling Large Files with Git LFS

Since a large dataset was used, Git LFS (Large File Storage) was implemented to manage it efficiently.

sudo apt install git-lfs  # Install Git LFS

This prevented upload errors and ensured smooth version control of large files.

---

What I Learned

This project provided valuable insights into:

✅ How recommendation systems work

✅ Working with APIs (TMDb) to fetch data

✅ Deploying a project for free with Streamlit

✅ Handling large datasets using Git LFS

✅ Building interactive UI with Streamlit

---

Try It Out!

The Movie Recommendation System is live and open for feedback!

🔗 Live Demo: Raso Movie Recommendation

🔗 GitHub Repo: Sourabh-Kumar04/Raso-Movie-Recommendation

🔗 Connect with Me on LinkedIn: Sourabh Kumar

---

Final Thoughts

Writing this first blog on Hashnode has been a great experience! 🚀

Let me know your thoughts in the comments. Also, feel free to connect and share feedback!

What should I build next?