Data analytics has undergone massive advances since the last UK General Election. Party campaigns will now be defined by their social media presence and the messages that they are pushing online, which means we need to find innovative ways to examine these new and rich data sources.
Nesta is working with the University of Sheffield to create the Political Futures Tracker, a tool that will provide near-real time analysis of political texts in the run up to the 2015 UK General election. It will hone in on key topics, future thinking and sentiment, as they unfold on Twitter, party websites and in political speeches and other relevant web material.
What will it do?
The Political Futures Tracker will enable us to analyse the levels of future thinking and sentiment in texts, as well as identifying key content and themes that politicians are writing and speaking about.
We feel that the extent to which politicians are thinking about the future matters. Political rhetoric is often focused on the short term, making political footballs out of important and often contentious topics. The political landscape in the UK encourages this short term outlook, but as research has shown, this sometimes means that politicians miss the next big thing.
How will it work?
We are exploring new ways to analyse a large amount of data from various sources. It is expected that both the amount of data and the speed that it is produced will increase dramatically the closer we get to election date. Using a semi-automatic approach, text analytics technology will sift through content and extract the relevant information. This will then be examined and analysed by the team at Nesta to enable delivery of key insights into hotly debated issues and the polarisation of political opinion around them.
The team at the University of Sheffield has extensive experience in the area of social media analytics and Natural Language Processing (NLP). Technical implementation has started already, firstly with data collection which includes following the Twitter accounts of existing MPs and political parties. Once party candidate lists become available, data harvesting will be expanded accordingly.
In parallel, we are customising the University of Sheffield’s General Architecture for Text Engineering (GATE); an open source text analytics tool, in order to identify sentiment-bearing and future thinking tweets, as well as key target topics within these.
One thing we’re particularly interested in is future thinking. We describe this as making statements concerning events or issues in the future. Given these measures and the views expressed by a certain person, we can model how forward thinking that person is in general, and on particular issues, also comparing this with other people. Sentiment, topics, and opinions will then be aggregated and tracked over time.
We will be publishing regular blogs on findings from the Political Futures Tracker, read them and find out more here.