Using NLP and Binary Classification for Predicting Suicidal Tendencies Based on Textual Input

Abstract

Suicide is death caused by injuring oneself with the intent to die. Suicidal tendency is a certain set of behavior that an individual exhibits when being suicidal, which includes posting cryptic messages in social media. This is a study that uses the Suicide and Depression Dataset and applies NLP and preprocessing techniques to transform the data. Machine learning models are then used to predict the class of the entries. Logistic Regression is the best performing model with an accuracy score of 93.24%. The best performing model is integrated and used in a Discord bot application.