How accurate do you think this can be? What happens if I write a facebook update that says, as an example: I don't believe I have ever said "I feel depressed" Would this be flagged? Does it only take one post to be flagged or is it looking for recurring behaviour?
If the last few posts leading up to the flagged post are classified as predominately negative (currently looking at the last 3 tweets before the flagged tweet), then we send the notification. The intuition is that if a person is comfortable enough with social to post seriously suicidal content, he or she has probably already made some preceding negative remarks.