🎬 Movie Recommender System (Content-Based)

Project Overview

This is a Content-Based Movie Recommender System built in Python. The system recommends movies similar to a user-selected movie based on title and genre metadata using TF-IDF vectorization and cosine similarity.

This project was created to demonstrate skills in Python, data preprocessing, machine learning, and web app deployment.

---

Features

• Content-based recommendation using TF-IDF and cosine similarity.
• Preprocessing includes cleaning titles and genres for accurate matching.
• Interactive web interface using Streamlit:
  • Select a movie from a dropdown
  • Choose the number of recommendations
  • Optional: Display movie posters from TMDB API
• Save/load model artifacts using joblib and numpy.
• Modular, well-structured project for easy extension.

---

Folder Structure

movie-recommender/ │ ├── app/ │ └── streamlit_app.py # Streamlit frontend ├── data/ │ ├── movies.csv # MovieLens dataset │ └── ratings.csv # Movie ratings ├── models/ # Saved model artifacts (after build) ├── notebooks/ │ └── exploration.ipynb # Data exploration and testing ├── tests/ │ └── test_recommender.py # Unit tests ├── recommender.py # Core recommendation logic ├── requirements.txt # Python dependencies └── README.md

yaml Copy code

---

Screenshots

1. Home Page / Movie Selection

2. Recommendations

---

Setup Instructions (Windows)

Follow these steps to run the project locally:

1. Clone Repository

git clone https://github.com/YOUR_USERNAME/movie-recommender.git
cd movie-recommender

2. Create Virtual Environment

Copy code
python -m venv venv
.\venv\Scripts\Activate.ps1
⚠️ If activation is blocked, run once:

powershell
Copy code
Set-ExecutionPolicy RemoteSigned -Scope CurrentUser

3. Install Dependencies

powershell
Copy code
pip install -r requirements.txt

4. Download Dataset

Go to MovieLens Small Dataset

Download and extract movies.csv and ratings.csv into the data/ folder

5. Build Model Artifacts

powershell
Copy code
python recommender.py
This creates TF-IDF vectors, similarity matrix, and title index in the models/ folder.

6. Run Streamlit App

powershell
Copy code
streamlit run app\streamlit_app.py
Open the URL printed in PowerShell (usually http://localhost:8501)

Select a movie and view recommendations

7. Run Unit Tests (Optional)

Copy code
python -m unittest discover -v
Libraries Used
Library	Purpose
pandas	Data manipulation, reading CSVs
numpy	Numerical operations, saving/loading arrays
scikit-learn	TF-IDF vectorization and cosine similarity
streamlit	Web interface for interactive recommendations
requests	Optional: Fetch movie posters from TMDB
joblib	Save and load model artifacts efficiently
unittest	Unit testing framework

Usage

Select a movie from the dropdown

Adjust the number of recommendations

Optional: Input your TMDB API key to show posters

Click "Get Recommendations" to see the top similar movies

Future Improvements

Include plot summaries, cast, or keywords for better similarity

Deploy online via Streamlit Cloud for portfolio showcase

References

Scikit-learn: TF-IDF Vectorizer

Streamlit Documentation

Author

Chidwan AD