Slashdot Mirror
Machine Learning System Detects Emotions and Suicidal Behavior
An anonymous reader writes with word as reported by The Stack of a new machine learning technology under development at the Technion-Israel Institute of Technology "which can identify emotion in text messages and email, such as sarcasm, irony and even antisocial or suicidal thoughts." Computer science student Eden Saig, the system's creator, explains that in text and email messages, many of the non-verbal cues (like facial expression) that we use to interpret language are missing. His software applies semantic analysis to those online communications and tries to figure out their emotional import and context by looking for word patterns (not just more superficial markers like emoticons or explicit labels like "[sarcasm]"), and can theoretically identify clues of threatening or self-destructive behavior.
More of the one size fits all BS. Sadly humans are not binary thus are a bitch to really understand their motivations.
Finally, I can use this
Can it tag a switch to Beta (either all at once or feature by feature over a year) as suicidal behavior?
Don't waste your vote! Vote for whoever you want, unless you live in a swing state it won't matter anyways
Machine Learning System can detect X with such high accuracy and with so few false positives as to be actually useful.
Frankly detecting X is basically defined as getting better than random chance. If you decide that anybody who posts "I am so sad" is suicidal, you'll bound to get a few hits, so there. I developed an algorithm that can detect suicide and depression. The problem here is that it's useless unless it's really really accurate.
It is no longer uncommon to be uncommon.
FTA: “Now, the system can recognise patterns that are either condescending or caring sentiments and can even send a text message to the user if the system thinks the post may be arrogant”
On the one hand, maybe it's a good idea to notify users that their comments will likely be interpreted by most readers as having 'X' emotional tone. On the other hand, it may result in people habitually self-censoring to the extent that they show no warning signs before they explode, (literally or figuratively), in some destructive action or activity.
I'm also thinking that this kind of ongoing **parentalistic monitoring is the wet dream of corporate overlords and wannabe dictators the world over.
--
**A word I coined, not a spelling mistake...
'The Economy' is a giant Ponzi scheme whose most pitiable suckers are the youngest among us and the yet-unborn.
Yeah - I looked for the paper that won him the Amdocs prize but couldn't find it. All reports seem to be, um, based on this story. Which is where I found he trained the system using two Fffacebook pages:
posts on Hebrew-language Facebook pages that are almost pure opinion, called “superior and condescending people” and “ordinary and sensible people.” The pages are basically forums for people to let off steam about things and events that get them mad, a substitute for actually confronting the offending person. Between them, the two pages have about 150,000 “likes,” and active traffic full of snarky, sarcastic, and sometimes sincere comments on politics, food, drivers, and much more.
“Now, the system can recognize patterns that are either condescending or caring sentiments and can even send a text message to the user if the system thinks the post may be arrogant,” explained Saig.
System Alert - Possible Arrogance Detected - user message issued
So it's a startup pitch - expect optimistic projections of outcomes. It's even possible (would it detect that) it's based on pure supposition - you know, like maybe the opinion of the machine learning program matched a readers take on those Fffacebook pages.
Oldie but goodie... Clippy and the suicide note.
Of course it can detect sarcasm, the algorithms required are really simple.
Get free satoshi (Bitcoin) and Dogecoins
On the heels of Google's "AI" that the WSJ claims got "testy" comes this claim of a "machine learning" system that can identify suicidal tendencies. Once again, BOGUS! The claim that machines learn anything is bogus to begin with, as to date no machine has ever done anything other than record information, as in so-called maze-learning programs. Learning is a cognitive process, and until we ourselves know how it works in humans (which we don't), we can never program a machine to learn anything.
But the real proof that this is bogus is that in order to "learn" to identify suicidal thoughts, even we humans would have to be given evidence that a given text actually came from someone as a result of their suicidal thinking. Which nobody can do, as this would require clairevoiance. Even the best psychiatric researchers can't know what someone was thinking when they composed a particular text.
AI is being dramatically overstated once more. What AI researcher has the guts to call them out on this?
Damn Auto Correct.
English... how about other's, like Hebrew, for example?
Or folks not writing in their native language on forums/social networks but in English which may be substantial.
Sounding sarcastic, critical, suicidal or otherwise emotional may not be authentic.
Not sure what the actual benefit would be using this and how many false positives this could create if some institution like DHS would use such a thing.
Yeah, that'll work.
I was thinking of "I Robot"
Interesting twist on the Turing test, can a program detect sarcasm in comments more accurately than humans?
Those four works are radically different, so I suppose between them they're bound to get a few things right.
To have a right to do a thing is not at all the same as to be right in doing it
Interesting twist on the Turing test, can a program detect sarcasm in comments more accurately than humans?
Yeah, sure it can.
To have a right to do a thing is not at all the same as to be right in doing it