Scroll the list beside to highlight the correlated paths on the chart. In order to filter the users, drag the mouse on the axis to select ranges of data.
Net Neutrality debate has been amplified a lot across social media, using the dedicated hashtag #netneutrality. In order to understand what people discussed about, what terms have been used and by who, we chose to observe twitter for two weeks—from Nov 25th to Dec 7th 2014—collecting all tweets containing the hashtag #netneutrality. Data have been obtained with the TCAT tool. 47852 tweets have been analyzed, with 7520 users involved (authors and\or mentions).
The graph shows the different patterns of the users involved in the debate, using three parameters: number of mentions, number of tweets posted and the number of followers. The parallel coordinates plot helps connecting these values in order to understand the different users patterns. The color of the paths changes from a cooler blue—meaning a low activity—to a warmer red, corresponding to a higher activity on twitter.
Scroll the list beside to highlight the correlated paths on the chart. In order to filter the users, drag the mouse on the axis to select ranges of data.
Users have been ordered by tweet activity, keeping only the top 500 results. Bot accounts have not been included in order to make the visualization more readable: some the most active users (@All4NeutralNet, @RealNeutralNet) are accounts which tweet automatically when someone signs up for a petition (about 1 tweet per minute), others (@Jeff88Ho) are bots who retweet everything with the #netneutrality hashtag.
After that, the cleaned dataset has been plotted using an interactive parallel coordinates graph, a D3.js-based visual toolkit developed by Kai Chang.
At first glance the graph reveals a common trend: most mentioned users are the ones who tweet less. @tomwheelerfcc and @barackobama have respectively 17411 and 12834 mentions, though in the two weeks recorded they never posted any tweet with #netneutrality hashtag. Combining the three axes it’s possible—generally speaking—to identify some more specific clusters of users, usually identified by their category: News accounts (@guardian, @repubblicait, @verge, @france24, @cnbc, @ariannahuff, @arstechnica, @gizmodo) have 0-25 mentions, 0-1 tweets and 500k-3M followers; members of the US Congress (all the accounts starting with @rep*) have more than 700 mentions, 0-30k followers, no tweets. In order to investigate what people discussed about when dealing with #netneutrality — see the protocol and the co-hashtag graphs — users have been divided into 3 groups: most active, most influent and the inbetweeners.
