Bot Sentinel is a free platform developed to detect and track trollbots and untrustworthy Twitter accounts. Bot Sentinel uses machine learning and artificial intelligence to classify suspicious Twitter accounts, and then we store those accounts in a database to track each account daily. We use the data we collect to understand how these accounts affect public discourse and how we can minimize their negative influence. Our team developed our machine learning model using 2500 ordinary Twitter accounts tweeting about politics, current news, and other various topics. We also used 2500 trollbot accounts we identified by examining the text of their tweets. In total, we used a combined 5,000,000 tweets from ordinary and trollbot accounts to build our model.
Propaganda bots are pieces of code that utilize Twitter API to automatically follow, tweet, or retweet other accounts bolstering a political agenda. Propaganda bots are designed to be polarizing and often promote content intended to be deceptive.
Propaganda trolls are persons that use emotions, lies, false accusations, and broken logic to undermine political discussions on social media platforms and forums. Some propaganda trolls are paid and often work for an organization with a network of propaganda trolls pushing a specific message.
Trollbot is a classification we created to describe human controlled accounts who exhibit troll-like behavior. Some of these accounts frequently retweet known propaganda and fake news accounts, and they engage in deceptive bot-like behavior. Other trollbot accounts target and harass specific Twitter accounts as part of a coordinated harassment campaign. Ideology, political affiliation, religious beliefs, geographic location, or frequency of tweets are not factors when determining the classification of a Twitter account.
Untrustworthy accounts are maintained by people that create and/or promote deceptive content intended to deceive the reader. Untrustworthy accounts are often retweeted by trollbot accounts to help spread the bogus material.
Unlike trollbot accounts, classifying untrustworthy accounts is a manual process. We review hundreds of tweets and retweets during the review process. If an account has a large number of followers and a high percentage of misleading and/or factually incorrect tweets, we might classify that account as untrustworthy.
Our machine learning model was developed to identify accounts that exhibit irregular tweet activity related to politics. The more you exhibit irregular tweet activity, the higher your trollbot score will be.
Trollbot is a classification we created to describe human controlled accounts who exhibit troll-like behavior. Some of these accounts frequently retweet known propaganda and fake news accounts, and they engage in deceptive bot-like behavior. Other trollbot accounts target and harass specific Twitter accounts as part of a coordinated harassment campaign. Ideology, political affiliation, religious beliefs, and geographic location are not factors when determining the classification of a Twitter account.
Accounts with a trollbot score of 75% and above are tracked by the Bot Sentinel platform. We use publicly available data that is available to anyone with access to Twitter's API.
Trollbot accounts are listed automatically when an account receives a trollbot score of 75% and above. Your Twitter handle is automatically removed from our database if your score drops below 75%. We do not manually remove Twitter handles from the trollbot listing on our website.
No, of course not.
Yes. No trollbot detection tool is 100% accurate. Periodically the Bot Sentinel algorithm will give false positives and false negatives. If an account is classified as a trollbot account, and you are positive it is a false positive, please wait a few days and analyze the account again. The same will apply to accounts you think are a false negative.
A: Our machine learning model and algorithm doesn‘t care if an account is a Conservative or Liberal, political affiliation isn’t a factor. There are Conservative accounts with low scores, and Liberal accounts with high scores.There are more Conservative accounts listed on our website because there are more Liberals using our service to check Conservative accounts. If there were more Conservatives checking Liberal accounts, the number of Liberal accounts listed would increase significantly.