Volume 17, No. 4, 2020
A Machine Learning Framework For Prognostication Of Suicidal Ideation And Detection In Online Social Content
Ed Gowhar Hafiz Wani , Samiah jan Nasti
Abstract
Every year, almost 800,000 people commit suicide. Suicide remains the second leading cause of death among a young generation with an overall suicide rate of 10.5 per 100,000 people. Suicide ideation is viewed as a tendency to end ones’ life ranging from depression, through a plan for a suicide attempt, to an intense preoccupation with self-destruction. Suicide ideation expressed in social media has an impact on language usage. Many at-risk individuals use social forum platforms to discuss their problems or get access to information on similar tasks. With the widespread emergence of mobile Internet technologies and online social networks, there is a growing tendency for people to talk about their suicide intentions in online communities. This online content could be helpful for detecting individuals’ intentions and their suicidal ideation. Some people, especially adolescents, choose to post their suicidal thoughts in social networks, ask about how to commit suicide in online communities, and enter into online suicide pacts. The anonymity of online communication also allows people to freely express the pressures and anxiety they suffer in the real world. Thus, one possible approach to preventing suicide effectively is early detection of suicidal ideation. In this research paper, we investigate the problem of suicidal ideation detection in online social websites, with a focus on understanding and detecting the suicidal thoughts in online user content (tweets/posts) and then apply machine learning techniques to automatically classify them into different categories of risk.
Pages: 44-49
Keywords: Suicide, Suicide Attempts, Suicide Ideation, Machine Learning.