The GOV.UK feedback form was receiving a lot of spam requests. We developed a machine learning model to detect spam responses — here is how we created it.
Discover how DSTL are able to share data with partners, by using synthetic data generation techniques to remove or obscure sensitive data while retaining its structure and characteristics
We recently held our first completely remote Data Science Community of Interest event. Hillary and James talk in this post about the event and recap some of the presentations
Read about how GOV.UK productionised the related links section, taking the outputs from a machine learning algorithm and adding them to the live workflow for publishers
This post describes how we test the performance of machine learning algorithms to improve navigation on GOV.UK
Read about Natural Language Processing projects happening in data science teams across government
GOV.UK can be represented as a multilayer network GOV.UK is a group of pages that comprises the best place to find government services and information. In its most abstract form, GOV.UK can be represented as a network made of pages, …
Find out all the ways in which semantic similarity is helping us improve GOV.UK
We're using supervised machine learning to organise all the content on GOV.UK, which means we can do things like create step by step journeys and consider voice activation. Here's what the data science team did.
One of the questions the data science team at the Government Digital Service (GDS) often gets asked is ‘how does your team work on data science projects?’ We thought we’d blog about how we have evolved our approach to data …