Detecting suicidality on social media: Machine learning at rescue

• Published in: Egyptian informatics journal 2023-07, Vol.24 (2), p.291-302
• Main Authors: Rabani, Syed Tanzeel, Ud Din Khanday, Akib Mohi, Khan, Qamar Rayees, Hajam, Umar Ayoub, Imran, Ali Shariq, Kastrati, Zenun
• Format: Article
• Language: English
• Subjects: Ensemble learning; Feature engineering; Informatik; Information Systems; Machine learning; Social media; Suicidal ideation
• ISSN: 1110-8665; 2090-4754; 2090-4754
• DOI: 10.1016/j.eij.2023.04.003

The rise in technological advancements and Social Networking Sites (SNS) made people more engaged in their virtual lives. Research has revealed that people feel more comfortable posting their feelings, including suicidal thoughts, on SNS than discussing them through face-to-face settings due to the social stigma associated with mental health. This research study aims to develop a multi-class machine learning classifier for identifying suicidal risk levels in social media posts. The proposed Enhanced Feature Engineering Approach for Suicidal Risk Identification (EFASRI) is used to extract features from a novel dataset collected from Twitter and Reddit platforms. Three machine learning algorithms, i.e. Support Vector Machine (SVM), Random Forest (RF) and Extreme Gradient Boosting (XGB) were employed for classification. The study demonstrates significant improvements in the precision, recall, and overall accuracy compared to previous research that used classical feature extraction mechanisms. The best-performing algorithm, Extreme Gradient Boosting (XGB), achieved an overall accuracy of 96.33%. The findings imply that different features contain different levels of information, and the right combination of the features supplied to the machine learning algorithms may improve the prediction results.