4 ms·
Sorry to tangent into the tech side of your post; but do you mind talking about what you use to do the filtering? I'm in a similar space, scraping content for
by existencebox 8y ago
Sorry to tangent into the tech side of your post; but do you mind talking about what you use to do the filtering? I'm in a similar space, scraping content for textual analysis alongside the images, but certainly run into the same problem. That being said, 8 billion images is a whole other level of scale to what I've touched thus far, and I'm curious how you've approached this problem, because that's CERTAINLY beyond a manual threshold.
- dumbfounder 8y agoIf I encounter truly objectionable content I remove a network of users and all of their content with a single click. It sometimes removes users that are innocent, but better to be safe than sorry. I also have a form on the site where you can report content, so I get help from visitors to identify the bad stuff. (Twicsy at its height had over 9 million visitors per month)