Social Media as a Mirror: Reflecting Mental Health Through Computational Linguistics

Abstract

The world is grappling with a serious problem: many young individuals are taking their own lives. There is also a challenge in understanding the rising trend of this tendency. It is essential to explore the reasons driving people of all ages to consider suicide and find ways to encourage them to choose life instead. In the modern era, social media acts as a crucial platform where people share their thoughts, activities, and emotional states. This has led to the consideration of whether analyzing social media posts could help discern whether individuals are experiencing joy or sadness, particularly to detect levels of sadness that could indicate suicidal thoughts. This paper employs artificial intelligence and machine learning tools to analyze the social media posts of individuals to gauge their mental state, specifically targeting signs that might indicate a risk of suicide. The study has found a high frequency of suicidal thoughts among those who appear depressed on social media. This research investigated the possibility to identify the likelihood of someone contemplating suicidal through their online behavior. This research demonstrates the potential of utilizing social media analysis to identify and support individuals at risk of suicide, providing new insights into recognizing and assessing suicidal thoughts and representing a significant advancement in suicide prevention efforts.

Keywords

• Depression
• exploratory analysis
• feature extraction
• LDA
• NLP
• suicide