Keep up with the news by installing RT’s extension for . Never miss a story with this clean and simple app that delivers the latest headlines to you.

 

Smart psychoanalysis: What your metadata can tell the NSA about you

Published time: March 31, 2014 03:59
AFP Photo / Michael Loccisano

AFP Photo / Michael Loccisano

Although the NSA is expected to cut down metadata collection, the telecom groups will still keep all the records. It seems like nothing personal, but the technical character of phone calls could surprise with the abundance of hidden material.

In a fact sheet published last week, the Obama administration pledges changes to the National Security Agency’s bulk telephony metadata program.

One of the expected concessions to civil society is the promise to leave records at the telephone companies, so that the government would allegedly be able to obtain them only in an emergency situation. But in the cold light of the day, records will still be kept. So, what could the ‘metadata’ –information on personal phone calls, claimed to contain no names or content – reveal to the NSA or just to the people who have access to them?

Social science has got a clue: and it’s not only the record itself, but also the technical information that could possibly expand the intrusion into private lives of people. A paper published in the Proceedings of the 6th International Conference on Social Computing, Behavioral-Cultural Modeling and Prediction shows that the data the NSA collects turns out to be of highly personal character.

A group of US and French researchers found that personality of the speaker on the phone could be reliably predicted by the way he or she makes calls. Their database was collected in the MIT Human Dynamics Lab at a leading US research university in a year-and-a-half-long study from March 2010 to June 2011.

The researchers, Yves-Alexandre de Montjoye, Jordi Quoidbach, Florent Robic and Alex Pentland used the 36 easy-to-obtain metadata elements, which fall into five categories:

Basic phone use including the number of calls;

Active user behaviors, as in the number of calls initiated, and the time it took the subject to answer a text;

Location, or how far the subject moved, the number of places from which calls have been made, and other indicators of so-called radius of gyration;

Regularity of calling routine;

Diversity, defined as the ratio between the subject’s total number of contacts and the relative frequency at which he or she interacts with them.

To figure out the results of the metadata analysis and to compare it with the personality, ‘hidden’ behind it, the scientists had determined psycho-types of 100 students, who participated in the survey (with a final sample of 69 people), and divided them also into five categories:

Extraversion – people who tend to seek stimulation in the company of others, who are outgoing and energetic;

Agreeableness – warm, compassionate and cooperative people;

Conscientiousness – self-disciplined and organized people, who are eager for success;

Neuroticism – people who experience unpleasant emotions more easily;

Openness – intellectually curious and creative ones, who are open to feelings.

Later on, a special machine-learning algorithm determined which set of technical characteristics referred to what kind of personal traits, one of the so-called Five factor model of personality.

“We let the algorithm determine the right mix,” de Montjoye said. “Each indicator is useful but is conditional on all the other indicators. That doesn’t mean each one is causal or that people who travel more are neurotic. Let’s say that the relationships between A and B are not linear, if you do a linear progression you see no relationship; you do a quadratic progression, you do see how A can predict B.”

With the accuracy of 49% to 63% scientists could be able to predict which personality type a person would belong to – that is 42% better than on average. No obvious correlations, just combination of behaviors and their frequency could predict most common personality types.

And although, according to de Montjoye, “We are unique in a way we use a mobile phone or telephone”, the model he developed alongside with his colleagues can’t name a method to make you less easy to define.

Taking into consideration 6 billion world mobile phone subscribers, this study provides telephone and marketing companies with a vast field to investigate implementation of targeted advertising, for example – undermining the necessity to conduct basic research, involving interviewing. Or, we could say, it gives the scientific community an unbelievable sample of people from all over the world. But these are relatively peaceful purposes.

Despite a reform that was called a ‘turning point’ by the NSA whistleblower Edward Snowden, the NSA still has access to ‘impersonal’ data that is not so innocuous as it seems.

Follow us

Follow us