Skip to content
Image 1 of 6

emoto

Visualizing the emotional online response to London 2012
http://emoto2012.org
emoto captured and visu­al­ised the global response around the London 2012 Olympic Games on Twitter in an inter­active online visu­al­isa­tion and phys­ical data sculpture. In this data art commission, I had the creative lead in producing a real-time social sentiment visualization, data journalistic analysis of the games as well as a physical data sculpture.
Created by Moritz Stefaner, Drew Hemment, Studio NAND. A FutureEverything project for the Cultural Olympiad programme and London 2012 Festival. Supported by Lexalytics, co-sponsored by GE and funded by Arts Council England and WE PLAY/Legacy Trust UK.

Diagram of the emoto pipeline: from the Games and social response through sentiment analysis to three visualization outputs

Real-time sentiment visualization

During the Games, we visualized the emotional online response in real-time in a web-based interactive application. Our custom infrastructure consumed the Twitter Streaming API, looking for tweets related to the Olympics and categorized these into sentiment categories (from “happy” to “angry” or “sad” messages). In addition, we detected topics of interest such as disciplines, athletes, nationalities, etc., so in the end we ended up with a real-time sentiment profile for topics relevant to the games. These sentiment profiles were visualized based on a unique origami-like visualization form, allowing easy comparison of how popular or controversial different topics were at a given time. In addition, incoming tweets were visualized in a stream-like message view, featuring an “inverse parallax” effect, where bigger (“more important”) tweets would float more slowly on top, while less cited or referred to tweets were flying by faster in the background.

Story fishing

As a complement to the fast-moving, ever-changing real-time popularity contest, we investigated various individual stories, such as the reception of the opening ceremony in the US and the UK, or the sentiment development of individual topics on the opening weekend.

Map of the UK plotting opening-ceremony tweet sentiment as colored bubbles sized by tweet volume
Map of the US plotting opening-ceremony tweet sentiment as colored bubbles sized by tweet volume

In this context, we developed a chart type we called “sentigraph”, which encodes sentiment in color and vertical position of a line, and at the same time the number of messages in line strength — it proved to be highly effective.

Sentigraph tracing tweet sentiment and volume for topics like NBC, Tom Daley and Michael Phelps, with annotated peaks

Of course, in the end, we used it to try and to summarize the audience reponse to the games in one l-o-o-o-o-ng timeline.

The full Olympics sentiment timeline: one long annotated sentigraph of Twitter mood across the Games

Data Sculpture

In many ways the crowning piece of the project, and a conceptual counterpoint to the ephemeral web activities, our data sculpture preserved the more than 12 million tweets in physical form. We had 17 plates CNC-milled — one for each day of the games — with a relief heatmap indicating the emotional highs and lows of each day. Overlay projections highlighted individual stories, and visitors could scroll through the most retweeted tweets per hour for each story using a control knob.

The tweets and topics displayed in the installation can be investigated in interactive heatmaps. Rollover the rows to see a tooltip display of the most retweeted tweet on the given topic at the respective point in time.

Related