Israeli researchers have developed an innovative technology for early detection of suicidal tendencies based on automatic text analysis of social network content
Suicide is the number one cause for young people under the age of 24 and a significant cause of death worldwide. Approximately one million suicides globally every year, of which about 500 happen in Israel.
Although psychological, psychiatric, and social help are effective tools in preventing suicides but are only applied in cases where the problem has been diagnosed and the person is receiving treatment.
It is important to recognize beforehand those people with suicidal tendencies. The task is difficult since medical information related to mental health is protected by confidentiality protocols, and at-risk people often do not seek help.
Currently, models for early detection of people at a real risk of committing suicide are based on traditional statistical methods that provide predictions that are about as accurate as chance-level predictions.
The researchers, Professor Roy Reichart from the Technion and his team along with Professor Christa Asterhan and Dr. Yaakov Ophir from the Hebrew University, have developed tools that enable the detection of people at-risk before they know it.
The researchers discovered that people with real suicidal tendencies rarely use explicitly alarming words in their posts (such as “death,” “kill” or “suicide”).
More often, they use negative descriptive words (”bad,” “worst”), curse words (“f*ing,” “bch”), expressions of emotional distress (“sad,” “hurt,” “cry,” “mad”), and descriptions of negative physiological states (“sick,” “pain,” “surgery,” “hospital”). People who do not have suicidal tendencies tend to express more positive emotions and experiences, and more references to religion and positive outlooks on life – a correlation that matches many studies that identified these factors as representing immunity to mental and emotional distress.
According to Dr. Ophir, the idea for the research was born following the tragic death of David-El Mizrachi, a 16-year-old who committed suicide because he was bullied online and in person. “It quickly became apparent that detecting suicidal tendencies early enough requires interdisciplinary research that includes researchers from different fields. That is how this multi-university and the multi-disciplinary group was formed.”
The detecting system combines natural language processing and machine learning with theoretical and analytical tools from the realm of psychology and psychiatry.
The researchers analyzed more than 80,000 posts written on Facebook by American adults, comparing their language usage patterns with their scores on a wide range of valid psychological indices.
“The power of the natural language processing-based algorithm lies in its ability to analyze enormous quantities of linguistic clues – something that humans are not able to do,” explained team member Refael Tikochinski. “In this project, we integrated cutting-edge attention-based neural network modeling for text representation, with layered neural networks for classification.”
The study was published in Natural