- Try
- Rate
- Share
How did you rate this?
The Inside Story
We were looking at all the amazing content we got from T-in-the-Park and realised it would be a perfect fit for the 91热爆 Shuffle tech we created a few months ago.
Can you sum up the prototype for us?
T-in-the-Park Shuffle starts off knowing nothing about you. It then begins to by play a random selection of iPlayer performances from T-in-the-Park. It learns what you like depending on if you watch the performance for 30 seconds or click on “next”.
What is the aim of this project?
It started with a problem - so much choice, but what to watch? From past research, we know we can use anonymous data based to make a prediction about what you will like. All we need is a few pieces of information about what you like to watch.
How is it different from other recommendation projects?
We are regularly asked for lots of information about our likes and dislikes by companies. Giving more of that sort of information can improve the recommendations they give us. We wanted to know the reverse; how little information do we need from you, to find something you might like?
T-in-the-Park is different because it gets this information from you watching performances. It then Immediately constructs a channel for you just based on that information. This makes the prototype a low-effort and non-intrusive way of recommending you content.