Skip to Content, Navigation, or Footer.
Thursday, Nov. 21
The Indiana Daily Student

New website analyzes political tweeting

IU study tracks Twitter for false content

As election season approaches, the usual backstabbing, rumor-spreading and slander of politicians can naturally be expected.

A new website created by IU information and computer scientists, Truthy.indiana.edu, plans to highlight such misinformation and smear tactics occurring through the popular social media outlet Twitter.

“The research goal is to study how ideas propagate through social media,” said Filippo Menczer, an associate professor of computer science and informatics in the IU School of Informatics and Computing. “A lot of things propagate in an organic way, but people also try to engineer this process hoping to generate traffic around a topic that may be a lie.”

The website analyzes the thousands of tweets occurring by the hour in search of keywords related to politics and sorts them into commonly-recurring themes, Menczer said.

These patterns, referred to as “memes,” are then examined to gather more information about their background and creation, in search of tweets being generated with false content or to mislead people.

Menczer said he hopes the website helps raise awareness that not everything read can be considered true, and the source must be considered. He also said such tools looking to keep politicians and activists responsible might motivate candidates to clean up campaigns.

“Political operatives of questionable ethics might be attracted to this technique of distributing whatever information they like through social media because it’s hard to get caught, and when you do, it’s often too late,” Menczer said. “The embarrassment of manipulating social media may be enough motivation, though.”

The information acquired through this Twitter research device is documented with visual diffusions of the tweets related to each subject and a live tweet feed on the Truthy website, Menczer said.

Even the individual words of tweets are analyzed to measure the mood of the tweet, including hostile, kind, confused, calm and other classifications. Such data
allows the researchers behind Truthy to show where the information is coming from and how it is being so highly dispersed.

Menczer said the ability to distribute information so quickly and in large quantities through Twitter can be dangerous.

“With enough traffic, Twitter themes can be displayed on search engines like Google, and people tend to trust the information they see on such sites as being fact, even if its not,” Menczer said.

Jacob Ratkiewicz and Mike Conover, Ph.D. students working on the project, said they agreed with the potential danger in social media.

“Twitter is resilient and self-organizing,” Conover said. “On Twitter, you have no central control or constraint for journalistic integrity. It’s decentralized with no accountability. This creates an environment where rumors can spread like wildfire.”

Ratkiewicz said the ability of information to spread quickly to so many people can cause misinformation.

“If something comes to you from your friends, you are much more likely to accept it as fact,” Ratkiewicz said.

The researchers encourage users to partake in the interactive aspects that allow people to mark information as “truthy” when they think it is false or is generated with an agenda in mind.

“It’s like Wikipedia,” Conover said. “If you get enough people contributing, usually you get a good estimate of where the truth is.”

Get stories like this in your inbox
Subscribe