ShirtHub is a recommendation engine built on top of a Rails app used to track our instructor's tee-shirt collection. Proof-of-concept project to demonstrate the implementation of artificial intelligence principles.
The different branches represent different jQuery styling for the splash page.
- Stored user and shirt data using SQLite
- Populated database with random users, each with a random collection of shirts
- Implemented Recommendable gem to compare users and suggest shirts based on those similarities
- Evaluated and stored user set similarities with Redis
- Handled simultaneous queries to Redis using Sidekiq gem
- Asynchronous loading and parallax using AJAX and jQuery
Gems
- sidekiq
- recommendable
- redis (in recommendable.rb initializer file)
- sqlite (development)
- pg (production)
- jquery-rails
Built by Julian Taub and Lucy Orbach in May 2015
Our instructor Ben Gross for his support. Steven Nunez for permission to build on his shirt app.
ShirtHub is MIT Licensed. See LICENSE for details.