multi-arm-bandit-movie-receommenders | MVP for demo bandit and recommenders

multi-arm-bandit-movie-receommenders is a CSS library. multi-arm-bandit-movie-receommenders has no bugs, it has no vulnerabilities, it has a Permissive License and it has low support. You can download it from GitLab.

It is started from a code sample in multi-arm bandit, then I think it would be cool to build a online multi-arm bandit system MVP with off-line machine learning algorithms. The goal here is to apply multi-arm bandit algorithm to measure the quality of a recommendation system. The quality of a recommender system is often measured by A/B testing experiments. However, A/B testing has traditionally been used to measure metrics in static features. what brought me to explore multi-armed bandits as an alternative to A/B testing in the scope of recommender systems. This project implemented the explore and explicit multi-arm bandit algorithm with some movie recommenders. These recommenders are based on the similarity of movies genre, overview description and users ratings. The models are mostly being trained and saved in the local disk except for the ALS spark recommender. You can explore the MVP by running docker on your machine.