We've got great news this week for nation-state employees tasked with using social media to spark a class war in previously stable democracies! Facebook is patenting technology to decide if its users are upper, middle or working class -- without even using the usual marker for social class: an individual's income (the patent considers this a benefit).
Facebook's patent plan for "Socioeconomic Group Classification Based on User Features" uses different data sources and qualifiers to determine whether a user is "working class," "middle class," or "upper class." It uses things like a user's home ownership status, education, number of gadgets owned, and how much they use the internet, among other factors. If you have one gadget and don't use the internet much, in Facebook's eyes you're probably a poor person.
Facebook's application says the algorithm is intended for use by "third parties to increase awareness about products or services to online system users." Examples given include corporations and charities.
(Score: 2) by nobu_the_bard on Thursday February 15 2018, @01:43PM
super informal but ...
Couple of easy methods I can think of. It's often a point of pride for people that own a home to mention this publicly, or otherwise publicly discuss it at some point. Where they spend most of their time asleep would likely also betray it if you have access to location data with time stamps; most people will be at home or at work at least 1/3 of the day or more, so you've got it narrowed down to just two places from just that. Access to financial records will also tell you pretty quick. Even access to who they have financial dealings with alone might be a good hint.
Likewise something that will probably come up in a public discussion at some point. Maybe even more proudly mentioned in some cases than home ownership, particularly among younger users. There's probably also resumes, graduation announcements, job listings, and other sources out there for these things. Their current and past debt status might also give some hints if you can compare it to their age.
Little bit trickier, but you could probably ballpark it based on education level and estimated technical aptitude with enough information. You might be able to draw further inferences from the number of devices they've synced to your social platform service or other services sharing information with you, the number of things on their home network, the number of sites reporting being accessed by unique devices from one IP address you have identified as relating to the user, etc etc.
Pssh! "Yes"