Do social media algorithms really undermine the ability to make decisions freely?

Have you ever watched a video or movie because YouTube or Netflix recommended you? Or did you add your friends to Facebook from the list of “People you may know”?

Also, how does Twitter determine which tweets to display at the top of the feed?

These platforms are driven by algorithms that rank and recommend content based on data.

Woodrow Herzog, a professor of law and computer science at Northeastern University in Boston, explains:

If you want to know when a social media company is manipulating you to disclose information or get more involved, the answer is always.

So, if you make decisions based on what these algorithms show, what does that mean for your ability to make decisions freely?

What we see is tuned for us

The algorithm is a digital recipe. A list of rules for achieving results using a set of materials. For tech companies, the result is usually to make money by persuading them to buy something or by continuing to scroll to see more ads.

The ingredients used are data that we provide through our online activities, either intentionally or otherwise. Whenever you like a post, watch a video, or buy something, it provides data that you can use to predict your next move.

These algorithms can affect us even if we are not aware of it. As the New York Times’ Rabbit Hole podcast is investigating, YouTube’s recommendation algorithms can lead viewers to increasingly extreme content, leading to a rapid online evolution.

Facebook’s news feed algorithm ranks content and maintains platform engagement. It can cause a phenomenon called “emotional transmission”. In this phenomenon, seeing a positive post means writing a positive post yourself, and seeing a negative post increases the likelihood of creating a negative post. However, the effect size is small in this survey.

Also, the so-called “dark pattern” is designed to trick us into spending more and sharing and spending more on websites like Amazon. These are website design tricks, such as hiding the unsubscribe button and showing the number of people who are buying the product you are looking at. just now.. They unknowingly move you towards the actions the site wants you to take.

You are profiling

Cambridge Analytica, the company involved in the largest Facebook data breach known to date, claimed to be able to profile your psychology based on your likes. These profiles can be used to target you in political advertising.

“Cookies” are small pieces of data that track us throughout our website. These are records of actions taken online (such as clicked links and visited pages) stored in the browser. When combined with data from multiple sources, such as large-scale hacks, this is called “data enrichment.” You can link your personal data, such as your email address, to other information, such as your education level.

These data are used regularly by technology companies such as Amazon and Facebook to create our profile and predict future behavior.

You are predicted

So how much of your behavior can be predicted by an algorithm based on your data?

In our survey, published last year in Nature Human Behavior, we investigated this question by looking at how much information your friends posted on social media contained information about you.

We used Twitter data to estimate how predictable people’s tweets were, using only friend data. Data from 8 or 9 friends proved to be sufficient to predict someone’s tweet, as if it were downloaded directly (well over 50% accuracy, see graph below). ).In fact, it gives you 95% of the potential prediction accuracy that machine learning algorithms can achieve. However From friend data.

Average predictability from a circle of close friends (blue line). A value of 50% means getting the next word correctly in half the time. Most people have a vocabulary of about 5,000 words, so it’s not an average feat. This curve shows how well the AI ​​algorithm can predict you from your friend’s data. Approximately 8-9 friends are sufficient to accurately predict your future posts as if the algorithm had accessed your own data (dashed line).
Bagrow, Liu, & Mitchell (2019)

Our result means that even #DeleteFacebook (which tended after the 2018 Cambridge Analytica scandal) may still be able to profile because of the remaining social ties. .. And that’s before we think about things about Facebook that make it so hard to remove anyway.

I also found that it is possible to create a profile for Non-user — So-called “shadow profile” — based on contacts on the platform. Even if you’ve never used Facebook, shadow profiles can still be created if your friends are using it.

On social media platforms such as Facebook and Twitter, privacy is no longer tied to individuals, but to the entire network.

Is there any more free will?Not perfect

But not all hope is lost. If you delete your account, the information contained in your social connections with your friends will become obsolete over time. Predictability gradually declines to lower levels, eventually restoring privacy and anonymity.

It may seem that the algorithm is reducing our own thinking ability, but it is not always the case. There is little evidence of the effectiveness of psychological profiling that affects voters.

Most importantly, humans are just as important when it comes to the role of humans in algorithms, such as in spreading (wrong) information. On Facebook, the degree of exposure to different perspectives is more closely related to social groups than the way news feeds present content. On Twitter, “fake news” can spread faster than it really is, but it’s mainly the people who spread it, not the bots.

Of course, content creators use the algorithms of social media platforms to promote their content on YouTube, Reddit, and other platforms, and vice versa.

After all, there are people under every algorithm. And we influence it as much as it affects the algorithm.

This article has been republished by The Conversation under a Creative Commons license. Please read the original article.

Do social media algorithms really undermine the ability to make decisions freely?

Source link Do social media algorithms really undermine the ability to make decisions freely?

Back to top button