Report: Extracting the Political Signal from the Online Noise: A workshop on mining online content and political networks

9980232816994487415Today we had an interdisciplinary workshop bringing together scholars from political science and data analytics interested in using data-analytics methods to extract information of interest from the mass of online sources now available about politics. The workshop was co-organised by Dr. James P. Cross of the Dublin European Institute and Dr. Derek Greene of the Insight Centre for Data Analytics

In recent years there has been an explosion in the availability of political texts online. Speeches, government records, press releases and party manifestos have all made the transition from print to digital text. Complementing these traditional political texts, new forms of political texts have also emerged online on political blogs, newspaper comment sections, and microblogging platforms like twitter. These texts, the relationships between them, and the online contexts in which they are found have the potential to provide new insight into a whole host of questions that have occupied political science as a discipline. Simultaneously, the methodological challenges associated with extracting useful insights from this mass of structured and unstructured data are considerable. Methodological developments in the fields such as data analytics, natural language processing and the analysis of social networks have proven useful in this regard. This half-day workshop brought together scholars from political science and data analytics in order to explore the potential cross-overs in these fields. In doing so it aimed to open up new collaboration possibilities and demonstrate the usefulness of new approaches to analysing political text online.

The workshop saw papers presented on using topic models to track the plenary agenda of the European Parliament and the Irish Dáil, and the discussion of the Greek far right in Parliament and online. Other papers were also presented on tracking amendments to legislative texts, the promises and pitfalls of sentiment analysis, and critical discourse analysis using automated methods.

On the basis of the conversation had, it looks like there is significant interest in further collaboration opportunities, so watch this space.

The workshop program is below. If you are interested in the paper do contact the author to ask for a copy!

Workshop program

Leave a Reply

Your email address will not be published. Required fields are marked *