Emotional Impact of Suicide on Active Witnesses: Predicting with Machine Learning

Abstract

Predicting the impact of suicide on incidental witnesses at an early stage helps to avert the possible side effect. When suicide is committed in public, incidental observers are left to grapple with it. In many cases, these incidental witnesses tend to experience the emotional side effect with time. In this study, we employed a Machine learning algorithms to predict the impact of suicide and suicidal attempt on incidental witnesses. This prediction was based on the accounts of suicide given by selected participants who have witnessed the act. The accounts, which was pre-processed into a corpus, were manually annotated with predefined emotion categories. While sadness emerged as the most salient emotional impact on the witnesses, fear was found as the lowest of the emotional impact on the witnesses. However, the machine learning prediction yielded highest in predicting depression with insignificant variations in the other emotional categories. This nonetheless shows that people who have witnessed suicide or suicidal attempts are inherently affected by some form of emotions that may require urgent attention to alleviate. By evaluating the performance of the Machine learning algorithms, the Support Vector Machine was superior, in terms its prediction, then the Multinomial Naïve Bayes algorithm. The outcome of the study contributes to the pool of research that sought to advocate the use of Machine Learning for predicting social phenomenon.