A team of European researchers has developed the US Election 2020 Web Monitor to track the public debate and the candidates’ performance on the campaign trail. Real-time analyses of online coverage based on the webLyzard Web intelligence platform reveal opinion leaders, shed light on events that impact the coverage and help compare stakeholder perceptions.
Try out the dashboard at us2020.weblyzard.com
Like no other presidency in recent history, Donald Trump’s period of office has polarized the electorate and the resulting news and social media coverage. Heated online discussions reflect the prevalence of filter bubbles and echo chambers in the global debate. Web intelligence technologies are uniquely positioned to analyze such phenomena. While the influence of social media is not new, in the 2016 and 2020 election cycles the Twitter postings of the candidates themselves have become key moments that shape the public debate for days and weeks. In this year’s presidential race, the social and economic impact of COVID-19 eclipses other issues of national or global importance. Recent events such as the White House coronavirus cluster introduce additional uncertainties, amplified by statements from both sides that keep fueling the online discussions.
The figures above show the results of a media analysis between Sep 23 and Oct 17, which is also available in PDF format. The analysis sheds light on the major stories emerging in conjunction with the Republican and Democratic campaigns, the top associations with each of the candidates as well as a comparison of Twitter frequency and sentiment – revealing the significant impact of the recent TV debates between the presidential candidates and their running mates.
Public Debate on the US Election 2020
How is this unusual presidential race perceived by citizens and stakeholders? To answer this question, the US Election 2020 Web Monitor analyses social media postings, stakeholder Web sites and the articles of international news media in English, French, Spanish, German and Dutch. Processing these multilingual content streams yields more than five million documents per month. Automated knowledge extraction services then classify and enrich the collected documents:
- They measure sentiment, for example, by associating candidate references with positive and negative expressions.
- Keywords grouped by political party, source or geographic region reflect events and emerging stories associated with the candidates.
- A visual dashboard provides a rich portfolio of interactive tools. This includes trend charts, word trees and geographic projections.
International Research Collaboration
The US Election 2020 Web Monitor is jointly pursued by researchers of webLyzard technology, the Modul University Vienna’s Department of New Media Technology and HTW Chur in Switzerland, who have a long history of successful collaboration. The underlying technologies and the visual dashboard are continuously being extended within the context of major EU research projects funded within the Horizon 2020 Programme (ReTV, EVOLVE and inDICEs).
Previous Awards and Presentations
In 2008, a monitoring application for the US presidential election won the “Online Communities, Web 2.0 and Social Networks” category of the Austrian National Award for Multimedia and e-Business. Eight years later, in June 2016, the US Embassy Vienna organized a joint presentation of Prof. Hans Noel from Georgetown University and webLyzard’s Managing Director, Prof Arno Scharl, to launch the US Election 2016 Web Monitor. In October 2016, the system featured in a TEDx talk on “Analyzing the Digital Talk: Visual Tools for Exploring Global Communication Flows” (see references below).
- Scharl, A. (2016). “Analyzing the Digital Talk: Visual Tools for Exploring Global Communication Flows“, TEDx talk at MODUL University Vienna in October 2016.
- Scharl, A. and Weichselbraun, A. (2008). “An Automated Approach to Investigating the Online Media Coverage of US Presidential Elections”, Journal of Information Technology & Politics, 5(1): 121-132.