News for hackers. Stuff that matters.
Re:After GDPR, The New York Times cut off ad exchanges and k... by ehnto on Wednesday January 16, 2019 @09:54AM (#18921293)
One thing I have wondered for a little while is how all these learning algorithms will handle change over time. A lot of my advertising is for things I may have liked perhaps 5 years ago, but no longer have anything to do with. It's also a bunch of super wrong things, like parenting, even though it also thinks I'm a single male who likes holidays, and skiing, even though it doesn't snow where I live and I've never even entertained the idea.

It's clearly confused, and so my wonder is how do they correct that over time? How do you un-scramble a scrambled machine learned profile? Can you even detect that it's gone awry?

Re:After GDPR, The New York Times cut off ad exchanges and k... by bryan_w on Wednesday January 16, 2019 @12:53PM (#18923069)
Yes! Both Facebook and Google have ad preferences pages you can go to to see what they have inferred about you and delete any wrong things
Re:After GDPR, The New York Times cut off ad exchanges and k... by rock_hard on Wednesday January 16, 2019 @12:53PM (#18923061)
The strongest signal to fix your targeting is to buy something you actually like from a ad!

The models react strongly to that...i went from years of poor ads to fantastic ads within 2-3 purchases