{"id":19537,"date":"2025-09-18T13:09:12","date_gmt":"2025-09-18T13:09:12","guid":{"rendered":"https:\/\/lonecandle.com\/?p=19537"},"modified":"2025-09-18T13:09:12","modified_gmt":"2025-09-18T13:09:12","slug":"analyzing-the-digital-traces-of-political-manipulationthe-2016-russian-interference-twitter-campaign","status":"publish","type":"post","link":"https:\/\/lonecandle.com\/?p=19537","title":{"rendered":"Analyzing the Digital Traces of Political Manipulation:The 2016 Russian Interference Twitter Campaign"},"content":{"rendered":"\n<p>\u201cWe collected a dataset with<\/p>\n\n\n\n<p>over 43 million elections-related posts shared on Twitter between<\/p>\n\n\n\n<p>September 16 and October 21, 2016 by about 5.7 million distinct<\/p>\n\n\n\n<p>users. This dataset included accounts associated with the identified<\/p>\n\n\n\n<p>Russian trolls. We use label propagation to infer the ideology of all<\/p>\n\n\n\n<p>users based on the news sources they shared. This method enables<\/p>\n\n\n\n<p>us to classify a large number of users as liberal or conservative<\/p>\n\n\n\n<p>with precision and recall above 90%. Conservatives retweeted Rus-<\/p>\n\n\n\n<p>sian trolls about 31 times more often than liberals and produced<\/p>\n\n\n\n<p>36 times more tweets. Additionally, most retweets of troll content<\/p>\n\n\n\n<p>originated from two Southern states: Tennessee and Texas. Using<\/p>\n\n\n\n<p>state-of-the-art bot detection techniques, we estimated that about<\/p>\n\n\n\n<p>4.9% and 6.2% of liberal and conservative users respectively were<\/p>\n\n\n\n<p>bots. Text analysis on the content shared by trolls reveals that they<\/p>\n\n\n\n<p>had a mostly conservative, pro-Trump agenda. Although an ide-<\/p>\n\n\n\n<p>ologically broad swath of Twitter users were exposed to Russian<\/p>\n\n\n\n<p>Trolls in the period leading up to the 2016 U.S. Presidential election,it was mainly conservatives who helped amplify their message.\u201d<\/p>\n\n\n\n<p><a href=\"https:\/\/arxiv.org\/pdf\/1802.04291\">https:\/\/arxiv.org\/pdf\/1802.04291<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"<p>\u201cWe collected a dataset with<\/p>\n<p>over 43 million elections-related posts shared on Twitter between<\/p>\n<p>September 16 and October 21, 2016 by about 5.7 million distinct<\/p>\n<p>users. This dataset included accounts associated with the identified<\/p>\n<p>Russian trolls. We use label propagation to infer the ideology of all<\/p>\n<p>users based on the news sources they shared. This method enables<\/p>\n<p>us to classify a large number of users as liberal or conservative<\/p>\n<p>with precision and recall above 90%. Conservatives retweeted Rus-<\/p>\n<p>sian trolls about 31 times more often than liberals and produced<\/p>\n<p>36 times more tweets. Additionally, most retweets of troll content<\/p>\n<p>originated from two Southern states: Tennessee and Texas. Using<\/p>\n<p>state-of-the-art bot detection techniques, we estimated that about<\/p>\n<p>4.9% and 6.2% of liberal and conservative users respectively were<\/p>\n<p>bots. Text analysis on the content shared by trolls reveals that they<\/p>\n<p>had a mostly conservative, pro-Trump agenda. Although an ide-<\/p>\n<p>ologically broad swath of Twitter users were exposed to Russian<\/p>\n<p>Trolls in the period leading up to the 2016 U.S. Presidential election,<\/p>\n<p>it was mainly conservatives who helped amplify their message.\u201d<\/p>\n<p>https:\/\/arxiv.org\/pdf\/1802.04291<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_monsterinsights_skip_tracking":false,"_monsterinsights_sitenote_active":false,"_monsterinsights_sitenote_note":"","_monsterinsights_sitenote_category":0,"footnotes":""},"categories":[13],"tags":[221,372,198,315,170,327,479],"class_list":["post-19537","post","type-post","status-publish","format-standard","hentry","category-article-share","tag-donald-trump","tag-election","tag-elections","tag-russia","tag-trump","tag-twitter","tag-voting"],"_links":{"self":[{"href":"https:\/\/lonecandle.com\/index.php?rest_route=\/wp\/v2\/posts\/19537","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/lonecandle.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/lonecandle.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/lonecandle.com\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/lonecandle.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=19537"}],"version-history":[{"count":1,"href":"https:\/\/lonecandle.com\/index.php?rest_route=\/wp\/v2\/posts\/19537\/revisions"}],"predecessor-version":[{"id":19538,"href":"https:\/\/lonecandle.com\/index.php?rest_route=\/wp\/v2\/posts\/19537\/revisions\/19538"}],"wp:attachment":[{"href":"https:\/\/lonecandle.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=19537"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/lonecandle.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=19537"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/lonecandle.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=19537"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}