Personalized web based application for movie recommendations
Duay, Killian (2019)
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2019060113994
https://urn.fi/URN:NBN:fi:amk-2019060113994
Tiivistelmä
Since a few years, the Machine Learning becomes more and more important. It is used everywhere, especially in recommender systems and personnalized marketing. Big companies of movies streaming build those systems in order to recommend some content to their customers and increase their profit.
The main issues with those companies is that they are focusing on their own content. They obviously only recommend the content they have and they are not taking into consideration the whole offer of existing movies. The user can not get global and objective recommendations because he is in the middle of several recommender systems that are not complete and not connected together since they are in competition.
The aim of this thesis is first to study the Machine Learning and the ways it can be used in the case of a movie recommender system. Then the aim is to implement one of those solutions in a project of a web based application for movie recommendations. The goal of the project is to develop a fully functionnal web application that allows users to get recommendations on all existing movies in the world. The application will be tested by real users.
The application will have a ReactJS frontend (web based and responsive design), a Python backend and a Firebase Database and Authentication. The movies and their information will be fetched on the TMDB API that provides data on all existing movies.
The deliverables are the fully functionnal project and this document. This document presents the background study, the design and implementation of the project, the results and the discussion about them and the possible further developments of the project.
The main issues with those companies is that they are focusing on their own content. They obviously only recommend the content they have and they are not taking into consideration the whole offer of existing movies. The user can not get global and objective recommendations because he is in the middle of several recommender systems that are not complete and not connected together since they are in competition.
The aim of this thesis is first to study the Machine Learning and the ways it can be used in the case of a movie recommender system. Then the aim is to implement one of those solutions in a project of a web based application for movie recommendations. The goal of the project is to develop a fully functionnal web application that allows users to get recommendations on all existing movies in the world. The application will be tested by real users.
The application will have a ReactJS frontend (web based and responsive design), a Python backend and a Firebase Database and Authentication. The movies and their information will be fetched on the TMDB API that provides data on all existing movies.
The deliverables are the fully functionnal project and this document. This document presents the background study, the design and implementation of the project, the results and the discussion about them and the possible further developments of the project.