Over the past week, Facebook was the center of controversy regarding its purported manipulation of the Trending Topics section.
At issue was a report published on tech blog Gizmodo alleging bias in the way Facebook manages Trending Topics results. Based on an internal document detailing “Trending Review Guidelines,” many users contend that Facebook purposefully manipulated the topics to exclude conservative news.
What are Trending Topics?
If you’re accessing Facebook through a browser, these topics traditionally appear in list form on the right-hand side of the window. They appear in a slightly different format on mobile devices.
The feature was added in 2014, separate from the News Feed, and offer a general topic list as well as topics based on politics, entertainment, sports and technology. Results are generated based on popular conversation topics and personalized for each user.
“Trending Topics is designed to help people discover major events and meaningful conversations,” said Justin Osofsky, vice president of Facebook’s global operations. “Topics that are eligible to appear ... are surfaced by our algorithms, not people.”
The problem with Osofsky’s last comment is that algorithms are codes generated by programmers, and programmers are people.
Anyone who knows programming knows algorithms can be manipulated to produce different results. Simply put, if a Facebook programmer changed something in the Trending Topics algorithm, the results would probably change.
The team managing the Trending feed uses a set of guidelines meant to ensure the product is “consistent with Facebook’s deep commitment to being a platform for people of all viewpoints,” Osofsky added.
“The guidelines demonstrate that we have a series of checks and balances in place to help surface the most important popular stories, regardless of where they fall on the ideological spectrum.”
According to the Facebook news site, here’s how the Trending Topics algorithm works:
The list of Trending Topics is personalized for each Facebook user.
This “personalization” algorithm relies on a number of factors, including the importance of the topic, pages and other content a user has liked, geographical location, feedback and what’s trending across Facebook overall.
Osofsky said, “Not everyone sees the same topics at the same time.”
Click on a topic and you move to a results page that includes all posts covering that topic, including relevant news sources. Just as the algorithm generates the list of trending topics, an algorithm generates the topic search results.
“Trending is also integrated into Facebook Search so you can search for any topic that may not show up in your Trending suggestions,” Osofsky noted.
Earlier in the week, Tom Stocky, Facebook’s vice president of search, said that the social media platform does not “insert stories artificially into Trending Topics.”
But Stocky’s comment actually contradicts Osofsky’s description of the role of the Trending Topics teams. For example, Facebook’s internal document shows three teams: editorial, topic detection and content ranking.
The very fact that teams oversee editorial control and rankings suggests Facebook is indeed attempting to serve a gatekeeper function.
Dr. Adam C. Earnheardt is special assistant to the provost and professor of communication in the department of communication at Youngstown State University in Youngstown, OH, USA where he also directs the graduate program in professional communication. He researches and writes on a variety of topics including communication technologies, relationships, and sports (with an emphasis on fandom). His work has appeared in Mahoning Matters as well as The Vindicator and Tribune-Chronicle newspapers.