A study, conducted by the University of Adelaide and the University of Vermont, claims that a person's behavior can be predicted using data from the social media of eight to nine of their friends.
The study, published in the journal Nature Human Behavior, found that deleting an account is not at all useful, as we can get information about a person from their friends' posts.
The researchers analyzed the content of about 30 million tweets, using information theory from mathematics and probability theory to test the predictability of individuals' behavior based on what they posted on social media.
The results showed that there was up to 95% accuracy in predicting someone's behavior from data taken from their friends alone. Data from eight to nine friends was enough to make predictions comparable to those based on the person's own data.
“There is no place to hide when it comes to social media platforms,” says one of the authors, Dr. Lewis Mitchell, a senior professor of applied mathematics at the University of Adelaide in the School of Mathematical Sciences and a research associate at the ARC Centre of Excellence in Mathematical and Statistical Frontiers.
"Telling people to delete their social media accounts to protect their privacy is not enough, as information such as political views or various interests can be determined from their friends' posts.
“Many people know they are giving away their information when they choose to use a social media platform, but they think it is only information that concerns them,” says Dr. James Bagrow, assistant professor of mathematics and statistics at the University of Vermont.
"But it's not an individual choice: they also give information about their friends.".
Dr. Mitchell added: “There are benefits to being able to predict behavior. platforms use this principle to get you posts that are of interest to you.”
“But of course there is a dark side, such as the potential for creating ‘filter bubbles’. For example, in a political debate, people may only be exposed to one type of information and not receive opposing views.”
