Algorithms in Social Media Platforms
How social media algorithms influence the spread of culture and information in the digital society.
By Maria Alessandra Golino
April 24, 2021
What is a social media algorithm?
Algorithms in social media platforms can be defined as technical means of sorting posts based on relevancy instead of publish time, in order to prioritize which content an user sees first according to the likelihood that they will actually engage with such content. For example, the posts which are recommended to you when you scroll through your Instagram feed, or the stories of your friends that appear first on the dashboard, are determined by algorithms.
Algorithms can be written by coders who make use of machine learning. “Machine learning” means that algorithms “learn” how to carry out tasks under various levels of human oversight. Algorithms manage several tasks which would be tedious for humans to carry out, such as managing flows of content through active recommendations as well as negative shadow bans and mediating interaction with information through likes and comments to improve content discoverability. In addition, algorithms rank and filter information in ways that create incentives and conditions of interaction for content creators that are similar to markets.
Why do algorithms exist?
The function of an algorithm is to deliver relevant content to users. The reason why social media platforms use algorithms is to more organically filter through the large amount of content that is available on each platform. Algorithms do the work of delivering content that is potentially more “interesting” for a user to the detriment of posts which are deemed irrelevant or low-quality - either in general, or to a specific user.
Regarding the criteria based on which algorithms deliver content, sometimes social media platforms explicitly specify which content they consider as high-quality and therefore promote on their platform. It must be further kept in mind that social media platforms are actual businesses, which make part of their revenue from marketing. This may include marketing of a brand or content that public pages want to promote by paying a fee to the social media in order to have the algorithms promote them.
How do algorithms work?
Algorithms are designed in a way that takes into account different aspects. Some of these aspects are content-based, meaning that this kind of algorithmic design seeks to match a user’s taste, based on their profile, to specific posts that the system guesses the user will like. Once users show interest in a specific tag or category, they are directed to other items in the same category.
Moreover, algorithms can operate in a collaborative way. Collaborative filtering consists in matching users to other users who seem to share similar interests; this way, a person is directed to posts or videos that they might want to see based on the fact that a user with a similar profile searched for that specific source. Algorithms can be context-aware, in the sense that they can individuate personal data such as a user’s exact geographic location in order to include it in the algorithmic calculations.
Finally, machine learning uses computers to simulate human learning, which allows them to identify and acquire knowledge from the real world, and improve performance of some tasks, such as recommendations through algorithms, based on this new knowledge.
How algorithms influence the spread of culture in society
These different approaches to algorithmic design have consequences when processing cultural content. For example, when engineers set computers through machine learning to create algorithms based on the geographic location of users, they limit - or at least direct - the spread of a particular form of art or of information to that specific area. The effects of algorithmic design can be considered both positive or negative. Often algorithms may be created with the aim of increasing awareness or interest in the digital society on a specific matter, some users may suddenly see in their feed an increase of posts concerning nutrition and diet, or foreign cinema, or politics.
However, the negative implications that algorithmic design may have are often the object of heated discussions on the controversies that surround algorithms. Such controversies often concern privacy issues: algorithms work with the personal data of the social media user, in order to “know” how to display the content on the social media platform (for example, algorithms make use of sensitive data such as the geographical location of the user, the friends and acquaintances they interact the most with, the pages and hashtags that they often search for, et cetera). Similarly, there are also considerations about how algorithms influence the opinion and interests of social media users and, consequently, of the digital society.
Through the use of shadow bans, algorithms may give rise to information gaps, as they hide or neglect certain posts, while prioritizing revenue-inducing content. This aspect of algorithmic design is controversial because it carries the pretense to determine which content users should find important or worth of appreciation. This may lead to a non-objective and polarized decision of who and what gets in the spotlight. As a result, algorithmic design inevitably influences the spread of culture and shapes the digital society in a certain way: it decides which type of content or topic should be given priority in each individual feed, and which artists, content creators or brands deserve to gain more visibility than others .
References & links
Hunt, R. & McKelvey, F. (2019). Algorithmic regulation in media and cultural policy: a framework to evaluate barriers to accountability. Journal of Information Policy 9, 307-335.
https://later.com/blog/instagram-shadowban/
Barnhard, B. (2021). Everything that you need to know about social media algorithms. Sproutsocial. Available at https://sproutsocial.com/insights/social-media-algorithms/
Poepsel, M. (2018). Digital Culture & Social Media in Media, Society, Culture & You. Rebus Community.
Maria Alessandra Golino
holds a Bachelor’s diploma in European Law from Maastricht University. Her role at the Institute for internet and just society is that of researcher for the Kittiwake project.
Read More
Watch Our Episodes