Review on Machine Learning Based Hybrid Model for Suicide Attempt Prediction

Review on Machine Learning Based Hybrid Model for Suicide Attempt Prediction

Sarika H. Zile¹, Dr. Smita L. Kasar², Daivashala R. Deshmukh³  

¹M.Tech. CST Student – G. S. Mandal’s Maharashtra Institute of Technology, Aurangabad, MH – IN

²Head of Dept. - CSE Dept – G.S. Mandal’s Maharashtra Institute of Technology, Aurangabad, MH – IN

³Asst. Professor - CSE Dept – G.S. Mandal’s Maharashtra Institute of Technology, Aurangabad, MH – IN

---------------------------------------------------------------------***---------------------------------------------------------------------

Abstract – Suicide attempts is a critical issue in modern society of the world. In this paper early detection and prediction of suicide attempts should be proposed to save people’s life. Machine learning techniques with feature engineering or deep learning for automatic detection based on online social contents and to perform awareness are currently suicidal information of detection methods. Clinical methods based on interaction between social workers or experts and the targeted individuals. The first survey that provides a comprehensive overview of these categories' methods and performance is included in this paper. The data sources of process applications of suicide attempts and detection—such as questionnaires, electronic health records, suicide notes, and online user content and datasets—are examined. In order to make it easier to continue the analysis of the research, numerous specific tasks and datasets are described and summarized. Finally, we offer a perspective on the proposed work's future research directions and a summary of the work's limitations.

Key Words: Suicidal ideation detection, social content, features engineering, Machine learning.