3 ms·
I think this sort of thing is dangerous because it's (probably) not data backed and you're really playing with dynamite. Do we know if this sort of message cau
by orik 10y ago
I think this sort of thing is dangerous because it's (probably) not data backed and you're really playing with dynamite.
Do we know if this sort of message causes a person to be more or less likely to take their lives? How can you A-B test that?
Does anyone else think your phone telling you it thinks 'you aren't being yourself anymore' could set someone over the edge?
- manufacturious 10y agoTo reply to your middle paragraph, you'd probably want a longitudinal cohort study rather than an A/B test. A well-designed one can provide pretty conclusive indications in situations where A/B testing (or, testing with a double blind control as the epi community might call it) isn't possible.
- chillacy 10y ago> Do we know if this sort of message causes a person to be more or less likely to take their lives? How can you A-B test that? Actually... you probably could train a model to detect suicidal posts. I had a friend who committed suicide a few months ago, and I didn't see it coming. But looking back, there were signs in his facebook posts (in line with this article's example of thanking everyone), but our conversations were normal. It is at the very least a wishful line of thinking for me that ML can recognize this type of pattern.