¡Hola!
The author of the project La Pelicula is on the line.
The project you are looking at right now was conceived as an experiment to test out a new technology announced at Microsoft Build 2025. When I heard about this, I thought it would be great to see it in practice. And, because I was thinking about building my own recommender system for quite some time, I thought that it is my chance. Thus, the project was born.
This is an example of a movie recommender system which you interact with (consciously or unconsciously) every time you open Netflix, Amazon Prime Video, Disney+, HBO Max or anything else (but probably a much simpler one).
From the very start, I wanted to keep as simple as possible (and, as a matter of fact, tried to remind myself about it along the development). So, basically, anyone can pick movie genres they like and get back a bunch of movies to watch based on their preferences. So, if you are struggling to choose which movie to watch tonight (with your friends, family or partner), then you have landed on the right page!
Now, going slightly more technical, it is based on a simple content-based neural network model which was trained using TensorFlow and Python. Neural Network was trained on the simplest dataset UI is implemented on top of Blazor, React, C# and JavaScript.
The data used for training is the MovieLens dataset from GroupLens Research, , a research group at the University of Minnesota.
The project is completely free to use and open-source. You can actually contribute to the project's source code if you'd like. However, as you might imagine, everything has a cost. I pay for the hosting and other things.
If you would like to support this project or, more general, what I am doing, you might buy me a coffee
Author of the project
