DeepHelp: Deep Learning for Shout Crisis Text Conversations

TL;DR

This paper develops deep learning models to analyze crisis text conversations, aiming to predict suicide risk, assess conversation success, and infer demographics, thereby enhancing mental health crisis intervention services.

Contribution

It introduces a modified Transformer-over-BERT model, a multitask learning framework, and a mathematical approach for bias correction in crisis text analysis.

Findings

Deep learning outperforms trained volunteers in suicide risk prediction.

Achieved 88.4% accuracy in predicting if a texter is 21 or under.

Bias correction suggests women are underrepresented in conversations.

Abstract

The Shout Crisis Text Line provides individuals undergoing mental health crises an opportunity to have an anonymous text message conversation with a trained Crisis Volunteer (CV). This project partners with Shout and its parent organisation, Mental Health Innovations, to explore the applications of Machine Learning in understanding Shout's conversations and improving its service. The overarching aim of this project is to develop a proof-of-concept model to demonstrate the potential of applying deep learning to crisis text messages. Specifically, this project aims to use deep learning to (1) predict an individual's risk of suicide or self-harm, (2) assess conversation success and CV skill using robust metrics, and (3) extrapolate demographic information from a texter survey to conversations where the texter did not complete the survey. To these ends, contributions to deep learning include a modified Transformer-over-BERT model; a framework for multitask learning to improve generalisation in the presence of sparse labels; and a mathematical model for using imperfect machine learning models to estimate population parameters from a biased training set. Key results include a deep learning model with likely better performance at predicting suicide risk than trained CVs and the ability to predict whether a texter is 21 or under with 88.4% accuracy. We produce three metrics for conversation success and evaluate the validity and usefulness for each. Finally, reversal of participation bias provides evidence that women, who make up 80.3% of conversations with an associated texter survey, make up closer to 73.5%- 74.8% of all conversations; and that if, after every conversation, the texter had shared whether they found their conversation helpful, affirmative answers would fall from 85.1% to 45.45% - 46.51%.