Movie Suggest

Try it here

...

About

A movie suggestion engine that recommends movies based on a movie that a user already watched and liked. Suggestions are based on the ratings given by other users.

It's based on the phrase: "People who liked this movie also liked this movie..."

Tech stack

Frontend

ReactJs with build deployed on Amazon S3 with cloudfront as CDN.

Backend

Dockerised Flask API with gunicorn as web interface deployed on Amazon elastic beanstalk.

Attributions

Data sets are provided by: F. Maxwell Harper and Joseph A. Konstan. 2015. The MovieLens Datasets: History and Context. ACM Transactions on Interactive Intelligent Systems (TiiS) 5, 4: 19:1–19:19. https://doi.org/10.1145/2827872

Movie images are provided by: The Movie Db https://www.themoviedb.org/

---

---

---

More info

The movie suggest engine inherently uses Collaborative Filtering a machine learning technique applied over real world user movie ratings.

The notebooks used for training the machine learning models used in this application are: https://www.kaggle.com/code/shashibheemanapally/movie-suggest-generate-x-collaborative?scriptVersionId=142716701
https://www.kaggle.com/code/shashibheemanapally/movie-suggest-cluster-centroids-st-deviation?scriptVersionId=143055433