Sans human, Facebook’s Trending Topics algorithm faired poorly

Posted to Mistaken Data  |  Tags: ,  |  Nathan Yau

Last week, Facebook announced that it was making the Trending Topics section more automated. More algorithm-based. Less person-based. On Monday, the section showed a fake news story at the top of the list for several hours.

Nick Statt for the Verge on the human element:

The changes instituted on Friday didn’t throw all of that away; Facebook has been slowly stripping away the human element of Trending Topics for months now. Rather, it marked the moment Facebook decided its algorithmic approach was more favorable, or perhaps more cost-effective and less damaging. But in shifting the reins to engineers, the company has minimized the kind of news judgment typically exercised by journalists and editors. Now, just a few days later, we’re realizing just how important that human element was.

Data. Always open for interpretation.

Favorites

Divorce Rates for Different Groups

We know when people usually get married. We know who never marries. Finally, it’s time to look at the other side: divorce and remarriage.

Visualizing the Uncertainty in Data

Data is an abstraction, and it’s impossible to encapsulate everything it represents in real life. So there is uncertainty. Here are ways to visualize the uncertainty.

Where Bars Outnumber Grocery Stores

A closer look at the age old question of where there are more bars than grocery stores, and vice versa.

10 Best Data Visualization Projects of 2017

It was a rough year, which brought about a lot of good work. Here are my favorite data visualization projects of the year.