How Facebook reveals family secrets and what a social network knows about us - ForumDaily
The article has been automatically translated into English by Google Translate from Russian and has not been edited.
Переклад цього матеріалу українською мовою з російської було автоматично здійснено сервісом Google Translate, без подальшого редагування тексту.
Bu məqalə Google Translate servisi vasitəsi ilə avtomatik olaraq rus dilindən azərbaycan dilinə tərcümə olunmuşdur. Bundan sonra mətn redaktə edilməmişdir.

How Facebook reveals family secrets and what the social network knows about us

Фото: Depositphotos

One day, Kashmir Hill was scrolling through his Facebook recommendations feed and came across a woman named Rebecca Porter. The surname seemed familiar to the journalist, and after contacting relatives, he found out that Rebecca was his second cousin, who lived in another part of the country, and whose existence he did not even suspect. But how then did the social network know what was unknown to Kashmir itself? The further Hill plunged into the mystery, the darker the picture of the omnipotence of Mark Zuckerberg's brainchild became. What else does the social network know about us? How does it collect and analyze information? Are the people she suggests in the “you may know them” section really unknown to us?

Portal «Idealist»translated the shocking investigation by Gizmodo.

Initially, Rebecca Porter and I were not familiar with each other. However, Facebook believed that we could be connected. Her name appeared this summer on my list “you can know them” - in this way the social network offers a selection of users you can add to yourself as a friend.

The function "You can know them" is known for its extraordinary ability to recognize those with whom you had a connection in real life. This puzzled and confused Facebook users when they found the name of an old boss in the list, a former one-night girlfriend, or someone they just encountered on the street.

Фото: Depositphotos

These friend suggestions go far beyond the usual communication of classmates or colleagues. Over the years, I have been told many strange stories about this, for example, when a psychiatrist told me that the social network recommends each other to her patients.

These recommendations are the result of various data sources: location information, activity in other applications, face recognition in photos. All these tools are at the disposal of Facebook, it can check the connection of its users with each other, hoping to engage their attention more and more. Users usually know that Facebook analyzes their online behavior, but the depth and perseverance of this monitoring is difficult to understand. And the function “You may know them”, or “PYMK” (People you may know) in the company's internal slang is a black box.

To try to understand this black box - an unknown and, apparently, aggressive data collection - I began to load and save a list of people recommended to me by the social network in order to try to understand the template.

Every day, the algorithm recommended to me about an 160 person, some over and over again; over the summer, the social network offered more than 1400 different people. About 200 people, or 15 percent of them, were in fact people I knew, but the rest, at first glance, were unfamiliar to me.

And then there was a story with Rebecca Porter. She appeared on the list in about a month: an elderly woman living in Ohio, with whom I had no friends on Facebook. I did not recognize her, but the last name seemed familiar. My biological grandfather, a man by the name of Porter, whom I had never met, left his father when he was a child. Father was adopted by a man by the name of Hill, and he did not know about his biological father until the age of majority.

The Porter family lived in Ohio. I grew up halfway across the country from them, in Florida, and knew about the existence of these blood ties, but did not think that someday I would meet these people.

A few years ago, my father finally met his biological father, as well as two uncles and paternal paternal aunt: his relatives sought him out when they came to the funeral of his mother in Ohio. None of them use Facebook. I asked my father if he recognized Rebecca Porter. He looked at her profile and said no.

I sent a message to Facebook to the woman, explained the situation and asked if she was connected to my biological grandfather.

“Yes,” she wrote.

Rebecca Porter, as we found out, turned out to be my second cousin on marriage. She met and married my biological grandfather 35 years ago, a year after my birth. It turned out that Facebook knew my family tree better than me.

“I didn’t know about you,” she told me when we talked on the phone. “I don’t understand how Facebook connected.”

It was a pleasant conversation. After finishing the phone call, I sat still for 15 minutes. I was grateful that Facebook gave me the opportunity to talk with an unknown relative so far, but this omniscience was frightening.

It was unclear how Facebook connected us. My father first met her husband after my grandmother’s funeral. They exchanged emails and the father recorded his number on his telephone. But none of them use Facebook. This also applies to other people who are between me and Rebecca Porter on the family tree.

It is known that Facebook is buying information from data brokers, and a person who previously worked for the company and is familiar with how this tool works suggested that the family connection may have been recognized that way. But in response to a question about this scenario, a Facebook representative stated: “the social network does not use information from data brokers for the“ You can know them ”function.

What information did Facebook use?

The company does not inform me of this, citing privacy concerns. A spokesman for the social network said that if the company responded to my request, it would have to answer similar questions to other users.

It was not a very convincing excuse. Facebook makes people send information about themselves all the time, why not share it back?

The main reason why the social network does not want to reveal how the recommendations work is that many of Facebook's competitors, such as LinkedIn and Twitter, offer similar functions to their users. In the 2010 presentation on PYMK, Facebook’s vice president of technology explained its value: “People with a lot of friends use the site more.” There is a competitive advantage that can be gained by being the best in this matter, and therefore Facebook is reluctant to disclose information about its algorithm.

The data tail is very large. Back in 2009, users who received incredibly accurate suggestions from friends suspected that Facebook bases its recommendations on the contact information that people provided when registering, not realizing that Facebook will save it and use it.

Despite the fact that Facebook now recognizes that it uses contact information, in 2009, the then Chief Privacy Officer Chris Kelly did not confirm this statement.

“We are constantly improving the algorithm we use to improve the“ Suggestions ”section on the home page,” Kelly said in an interview with Adweek in 2009. "But we do not give details about the algorithm itself."

The silence about the information about how this tool works, disappoints users who want to understand the degree of Facebook's awareness of them and how deeply the social network digs into their lives. A spokeswoman said that the recommendations of friends are formed thanks to more 100 signals, and one signal alone will not offer you a friend.

One hundred signals! I told a company representative that there could be greater transparency in the interests of the search giant regarding how this feature works so that users are less afraid of it. She replied that Facebook “in the name of transparency” recently added additional information to the help page explaining how the “you can know them” function works.

This help page offers a short bulleted list:

The recommendations in the “You Can Know Them” section are based on:

  • that you have mutual friends, including mutual ones (the most common reason for recommendations);
  • that you are in the same group on Facebook or have been tagged in one photo;
  • your communities (for example, school, university or work);
  • your downloaded contacts.

What about the other signals?

“We decided to list the most common reasons why someone might be offered as people you might know,” a Facebook representative said in an email when asked about the short list.

Instead of explaining how Facebook linked me to my second cousin grandma, a spokesman suggested that I e-mail me to delete this offer if I don’t like it.

“People don’t always like some of the PYMK offers, so there is an action that people can take to influence the algorithm — click on the cross next to the offer that doesn’t interest you,” the social network spokesman wrote to me by email the best way to tell us that they are not interested in a given person, and this feedback helps to improve our proposals over time. ”

Now, when I look at the recommendations, I am annoyed by not just seeing the names of people I know without it, but also all the alleged strangers on the list. I wonder how many of them are truly strangers, and how many are connected with me in such a way as I do not know. Perhaps these are the people whom I should know?

 

Read also on ForumDaily:

Big brother is here and his name is Facebook

How Russians seek their ancestors of the era of Ivan the Terrible and find relatives

Russian found great-grandfather executioners through 78 years after the execution

How an Israeli startup connects families to any part of the world

Miscellanea In the U.S. list relatives Facebook Social Network
Subscribe to ForumDaily on Google News

Do you want more important and interesting news about life in the USA and immigration to America? — support us donate! Also subscribe to our page Facebook. Select the “Priority in display” option and read us first. Also, don't forget to subscribe to our РєР ° РЅР ° Р »РІ Telegram  and Instagram- there is a lot of interesting things there. And join thousands of readers ForumDaily New York — there you will find a lot of interesting and positive information about life in the metropolis. 



 
1082 requests in 1,240 seconds.