Throughout the spring and summer, the Kremlin-backed troll account @TEN_GOP pushed political hashtags for divisive events, including the Charlottesville "Unite The Right" rally, according to a new analysis of Twitter data.



Last week, reports emerged that @TEN_GOP, a popular Twitter account that the Tennessee Republican Party purportedly managed, was, in fact, run by the Kremlin-backed Internet Research Agency, which operated a network of propaganda-pushing troll accounts. @TEN_GOP, which had over 140,000 followers, was widely cited in the pro-Trump media sphere and even retweeted by members of the Trump campaign just before the election.

Reports show that the account tweeted Russian propaganda, as well as polarizing political content, but since Twitter suspended @TEN_GOP in August (Twitter does not comment on individual accounts for security reasons), it's difficult to get a sense of what the account was tweeting, as well as the size of its influence.

However, a hashtag analysis by multimedia artist Erin Gallagher offers a glimpse at how the troll account worked in tandem with the pro-Trump media ecosystem to help push politically divisive narratives across Twitter. Gallagher, who maps prominent political and pro-Trump hashtags, noticed @TEN_GOP had appeared prominently in some of her old charts after last week's reports linking the account to the Internet Research Agency.

Below are four charts that show how @TEN_GOP helped spread hashtags related to some of the spring and summer's most politically charged events.

A few things to note before the data:

- The charts capture only four specific hashtags from the highly active account and thus reflect a very small portion of its total body of tweets.

- The size of the nodes demonstrates the reach of a specific account in spreading a particular hashtag (the bigger the node, the more accounts it reached) and the distance between nodes is an indication of degrees of separation of the accounts (the greater the distance, the less common connections the two accounts have).

- Gallagher cautions that the charts are best read as a snapshot of a moment at the time they were analyzed, meaning any tweet in this ecosystem could go viral and change the scale of the chart.

- The charts don't take into account sentiment information, meaning that the tweets about the event could be in favor of the event, against it, or neutral.

As such, Gallagher suggests that determining the precise influence of the account on its followers is both subjective and abstract. "It definitely had influence," she told BuzzFeed News, "but I'd say it was just one component in a big machine."

April 15, 2017 - #Berkeley Protests