Collective Intelligence for Suicide Surveillance in Web Forums

Abstract

Internet users may see and comment on suicide expressions in the cyberspace that are not identified by helping professionals. The communication between the two parties is not well facilitated. In light of this situation, we present a system using collective intelligence from Internet users to efficiently and effectively identify suicidal people in order to provide timely intervention and promote better public health. We describe the system architecture involving information retrieval, affect analysis, and opinion summarization technique. The collective intelligence approach incorporates machine learning techniques and Internet users' contributions to facilitate the automated identification of suicide expressions. The system will be examined and evaluated in lab settings and suicide prevention organizations. © 2013 Springer-Verlag.