3 ms·

If somebody wants to definitively answer this, there's no reason you couldn't just say random advertising related words in front of your phone (1 per day) and c

If somebody wants to definitively answer this, there's no reason you couldn't just say random advertising related words in front of your phone (1 per day) and check how many come up in your search. There are basic statistical methods to establish if the results can be explained by chance. This would have the upside of not requiring any reverse-engineering.

remir 9y ago
Yeah. For example, if you don't own a cat, you could just talk about cats and brands of cat food in front of your phone for a while. If you end up with a bunch of ads for cat products, then that would be seriously weird.
- madeofpalk 9y ago
  In saying that, Twitter thinks I: * Own a cat, dog and other animal * Have between $100k- $999k liquid investible assets * Have a net worth between $1 and $1m * Am highly affluent * Am a high spender * Am a frugal spender * Own a house * Have multiple families Yet none of these are true (well, I guess apart from the 'has income' demographic I'm in). I know Twitter isn't known for being an advertising powerhouse (esp. compared to Facebook), but I wouldn't take too much stock in Facebook serving me up irrelevant ads. I moved to the other side of the world 2 months, updated my "living" location on Facebook and have been tagged at multiple locations in my new city, yet Facebook still serves me ads for buying an iPhone or Car back in my home town. https://imgur.com/a/HIkic https://imgur.com/a/HIkic
delroth 9y ago
Wouldn't this still be subject to confirmation bias? As in: you'd be more likely to notice advertisement that is related to a word you pronounced recently, since you would be training your brain to look carefully for these words.
ng-user 9y ago
I don't have a link right now (youtube though) but I saw a couple do this with 'cat food' and it ultimately resulted in a brigade of ads despite not owning a cat.
- nelsonic 9y ago
  @ng-user The "cat food" couple video (linked to from the Wired Article: https://www.wired.com/story/facebooks-listening-smartphone-microphone https://www.wired.com/story/facebooks-listening-smartphone-m... ...) is: https://www.youtube.com/watch?v=U0SOxb_Lfps https://www.youtube.com/watch?v=U0SOxb_Lfps

remir 9y ago

Yeah. For example, if you don't own a cat, you could just talk about cats and brands of cat food in front of your phone for a while. If you end up with a bunch of ads for cat products, then that would be seriously weird.

madeofpalk 9y ago
In saying that, Twitter thinks I: * Own a cat, dog and other animal * Have between $100k- $999k liquid investible assets * Have a net worth between $1 and $1m * Am highly affluent * Am a high spender * Am a frugal spender * Own a house * Have multiple families Yet none of these are true (well, I guess apart from the 'has income' demographic I'm in). I know Twitter isn't known for being an advertising powerhouse (esp. compared to Facebook), but I wouldn't take too much stock in Facebook serving me up irrelevant ads. I moved to the other side of the world 2 months, updated my "living" location on Facebook and have been tagged at multiple locations in my new city, yet Facebook still serves me ads for buying an iPhone or Car back in my home town. https://imgur.com/a/HIkic https://imgur.com/a/HIkic

delroth 9y ago

Wouldn't this still be subject to confirmation bias? As in: you'd be more likely to notice advertisement that is related to a word you pronounced recently, since you would be training your brain to look carefully for these words.

ng-user 9y ago

I don't have a link right now (youtube though) but I saw a couple do this with 'cat food' and it ultimately resulted in a brigade of ads despite not owning a cat.

nelsonic 9y ago
@ng-user The "cat food" couple video (linked to from the Wired Article: https://www.wired.com/story/facebooks-listening-smartphone-microphone https://www.wired.com/story/facebooks-listening-smartphone-m... ...) is: https://www.youtube.com/watch?v=U0SOxb_Lfps https://www.youtube.com/watch?v=U0SOxb_Lfps