Natalia Bielacka, Marek Drwal, Piotr Bereznowski, Michał Kruczkowski
The aim of this project is to create a content-based Entertainment Recommender System. It would allow the user to browse catalogues of books, movies and games using various filters. While browsing media will be sorted by the rating and chosen characteristics. Media can be browsed by title (or part of it), genre, type of media or related people.
The system will also contain users’ opinions that all the users could browse. A user can add their own opinions and manage them. The rating of the specific media will be determined by all the opinions. Recommendation will be based on accumulated average rating and appropriate characteristics. There will be two types of recommendations – to specific media and to search by keyword and/or filters. Application will also allow to browse only chosen type of media via selecting dedicated subsection.
Scalability is fulfilled by:
- providing appropriate number of users (people interested in movies, books and games)
- suitable data size
- application of parallel processing algorithms
Potentially exploited gigabyte-large datasets:
- Amazon books reviews
- Goodreads book datasets
- Game recommendations on Steam
- Netflix prize data
- The movies dataset
- Inspired by Scrum-based workflow
- Primary communication medium: MS Teams (continuous contact)
- Weekly in-person meetings
- Job assignment via MS Teams add-on and for bigger tasks via GitLab Issues.