Abstract
Recent statistics in suicide prevention show that people are increasingly posting their last words online and with the unprecedented availability of textual data from social media platforms researchers have the opportunity to analyse such data. Furthermore, psychological studies have shown that our state of mind can manifest itself in the linguistic features we use to communicate. In this article, we investigate whether it is possible to automatically identify suicide notes from other types of social media blogs in two document-level classification tasks. The first task aims to identify suicide notes from depressed and blog posts in a balanced dataset, whilst the second experiment looks at how well suicide notes can be classified when there is a vast amount of neutral text data, which makes the task more applicable to real-world scenarios. Furthermore, we perform a linguistic analysis using LIWC (Linguistic Inquiry and Word Count). We present a learning model for modelling long sequences in two experiment series. We achieve an f1-score of <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">88.26</i> percent over the baselines of <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0.60</i> in experiment 1 and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">96.1</i> percent over the baseline in experiment 2. Finally, we show through visualisations which features the learning model identifies, these include emotions such as love and personal pronouns.
Highlights
W HILST both machine and deep learning techniques have been predominantly used for commercial purposes, there has been an increased awareness of how AI approaches could contribute to solving some of the biggest social problems humans face worldwide [1]. This awareness has led to the creation of new workshops and conferences that fall under the umbrella of AI for Social Good, where machine learning researchers connect with Non-Governmental Organisations (NGOs), charities and other problem owners to create practical solutions
Other work conducted by [31] has found that using a combination of both linguistic and sentiment features achieves an accuracy of 86.61% by using a logistic model tree (LMT)
Work on identifying depression and other mental health conditions has become more prevalent over recent years, where a shared task was dedicated to distinguishing depression and Post Traumatic Stress Disorder (PTSD) on Twitter using machine learning [9]. [40] have argued that changes in the cognition of people with depression can lead to different language usage, which manifests itself in the use of specific linguistic features
Summary
W HILST both machine and deep learning techniques have been predominantly used for commercial purposes, there has been an increased awareness of how AI approaches could contribute to solving some of the biggest social problems humans face worldwide [1] This awareness has led to the creation of new workshops and conferences that fall under the umbrella of AI for Social Good, where machine learning researchers connect with Non-Governmental Organisations (NGOs), charities and other problem owners to create practical solutions. Whilst there are efforts to tackle suicide and other mental health conditions online by social media platforms such as Facebook [7], there are still concerns that there is not enough support and protection, especially for younger users [8] Taking these trends into account and with this unprecedented availability of textual data from social media platforms researchers have the opportunity to analyse such data and use their findings in several different application areas.
Sign-in/Register to view highlights & summary 
Accepted Version (
Free)
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call