How to do Data Science when you don’t have any data…

This was exactly the situation that our Data Science Team faced during the Covid-19 crisis in March 2020. Find out how we created an ML quiz engine with no data or questions.

To find out how we did, try it for yourself!

The Challenge

BBC Bitesize wanted a 20-question quiz that helps GCSE students identify the areas that should be prioritised during their home learning.

We had no questions and no data to work with. On top of that, we needed to be confident about the results — we didn't want to direct students away from content that would be beneficial to their learning.

In order to secure the resourcing to create this new offering, we needed to create an algorithm that could deliver the level of adaptation and intelligence that stakeholders wanted. Creating an unintelligent data-collection quiz first was not an option, it had to be clever from the point of conception. However, any algorithm we created was always going to be inferior to one that is informed by data collected once the product goes live. We needed an algorithm that is intelligent without data, and flexible enough to…