Movies Recomendation using Machine Learning

Introduction

A movie recommendation system, or a movie recommender system, is an ML-based approach to filtering or predicting the users’ film preferences based on their past choices and behavior. It’s an advanced filtration mechanism that predicts the possible movie choices of the concerned user and their preferences towards a domain-specific item, aka movie.

About this project

This is a project for the SoyHenry academy.

Repository content

• etl.ipynb: Contain the ETL process for the CSV files and create a clean CSV.
• movies.py: Contain the class to process api requests.
• api.py: Contain the GET method to process server requests.
• eda.ipynb: Contain the EDA process and recomendation system.
• recommendation_system.py: Contain the machine learning recomendation system.

Workflow

For the project is used Github Action Project, you can access this project in the following link.
This project will follow the conventional commits of the following link.
Each task will be divided into an issue. Here you can find all the issues available at the moment.
The scheduled task timetable can be access in the following link.

The following image shows the workflow proposed for this project:

Data Engineering ETL

For this task it has been created the following issue where you can find the description of the task to solve.

API development

For this task it has created the following issue where you can find the description of the task to solve.
It has the following related issue.

API Endpoints

To access the endpoints you should use the following path:

https://moviesrecomendation.onrender.com/

• /cantidad_filmaciones_mes/{month}: Get the amount of movies released in the requested month.
• /cantidad_filmaciones_dia{day}: Get the amount of movies released in the requested day.
• /score_titulo/{title}: Get the released year and the score of the requested title.
• /votos_titulo/{title}: Get the released year, vote count and vote average of the requested title.
• /get_actor/{actor}: Get the amount of movies starred by the actor, the return of the movies and average return of the movie.
• /get_director/{director}: Get a list of movies of the director, the return, the budget and the revenues of each movie as a list.
• /recomendacion/{title}: Get the top 5 movies recommended based on overviews, gendre and score of movies.

Example:

Recommendation

https://moviesrecomendation.onrender.com/recomendacion/The Empire Strikes Back

Movies launched in the selected day

https://moviesrecomendation.onrender.com/cantidad_filmaciones_dia/lunes

Note: The endpoint are not key sensitive, that means that you can use lowercase or uppercase.

Deployment

To deploy the environment it has used the following tutorial. The related issue can be found here.

Exploratory data analysis

For this task it has been created the following issue where you can find all the related information.

Recommendation system

For this task it has been created the following issue where you can find all the related information.
You can access the server with the following link: