An interactive recommender system using a novel, drag-and-drop user interface and a fast, client-side recommender.
Project from 2012 - present
What we are doing
Sibyl is a fast, interactive recommender system that helps you find interesting TV and radio programmes. It builds on our previous experience of personalisation and interactivity gained in the project and the recommender system technology developed in the project.
Why it matters
There are so many programmes to choose from these days it can be hard to decide what to watch. Personalised recommender systems can help but usually require knowledge of your previous viewing history or preferences.
Unfortunately, some users may be new to our services, may not log-in or may not want their viewing history to be tracked. For these reasons we have been exploring techniques that do not depend on historical viewing records.
Our Goals
- to help people find interesting TV and radio programmes
- to explore alternatives to complex, centralised recommender systems
- to elicit user feedback in the easiest way possible for the user
Outcomes
has been released on the R&D Labs website, providing an opportunity to prove the effectiveness of client-side recommender system.
We've gained feedback from users which has allowed us to refine and improve the user interface concept and plan future work on recommender systems.
How it works
Sibyl generates recommendations based on feedback users provide about familiar programmes using a novel drag-and-drop user interface. It uses a fast, client-side recommender system implemented in JavaScript to reduce the latency associated with server-side systems.
This approach also has the advantage of being highly scalable and helps to maintain user privacy, since no user data is returned to our servers.
The underlying model for the recommender system is built using the Recommender System Library.