Abstract:Background: Suicide has been one of the leading causes of deaths in the United States. One major cause of suicide is psychiatric stressors. The detection of psychiatric stressors in an at risk population will facilitate the early prevention of suicidal behaviors and suicide. In recent years, the widespread popularity and real-time information sharing flow of social media allow potential early intervention in a large-scale population. However, few automated approaches have been proposed to extract psychiatric stressors from Twitter. The goal of this study was to investigate techniques for recognizing suicide related psychiatric stressors from Twitter using deep learning based methods and transfer learning strategy which leverages an existing annotation dataset from clinical text. Methods: First, a dataset of suicide-related tweets was collected from Twitter streaming data with a multiple-step pipeline including keyword-based retrieving, filtering and further refining using an automated binary classifier. Specifically, a convolutional neural networks (CNN) based algorithm was used to build the binary classifier. Next, psychiatric stressors were annotated in the suicide-related tweets. The stressor recognition problem is conceptualized as a typical named entity recognition (NER) task and tackled using recurrent neural networks (RNN) based methods. Moreover, to reduce the annotation cost and improve the performance, transfer learning strategy was adopted by leveraging existing annotation from clinical text. Results & conclusions: To our best knowledge, this is the first effort to extract psychiatric stressors from Twitter data using deep learning based approaches. Comparison to traditional machine learning algorithms shows the superiority of deep learning based approaches. CNN is leading the performance at identifying suicide-related tweets with a precision of 78% and an F-1 measure of 83%, outperforming Support Vector Machine (SVM), Extra Trees (ET), etc. RNN based psychiatric stressors recognition obtains the best F-1 measure of 53.25% by exact match and 67. 94% by inexact match, outperforming Conditional Random Fields (CRF). Moreover, transfer learning from clinical notes for the Twitter corpus outperforms the training with Twitter corpus only with an F-1 measure of 54.9% by exact match. The results indicate the advantages of deep learning based methods for the automated stressors recognition from social media.

Detecting and Explaining Crisis

Let's explain crisis: deep multi-scale hierarchical attention framework for crisis-task identification

Conceptualizing Suicidal Behavior: Utilizing Explanations of Predicted Outcomes to Analyze Longitudinal Social Media Data

Early Detection of Mental Health Crises through Artifical-Intelligence-Powered Social Media Analysis: A Prospective Observational Study

Early Detection of Mental Health Crises through AI-Powered Social Media Analysis: A Prospective Observational Study

Classifying Crises-Information Relevancy with Semantics

CrisisKAN: Knowledge-infused and Explainable Multimodal Attention Network for Crisis Event Classification

The Ethical Risks of Analyzing Crisis Events on Social Media with Machine Learning

An Annotated Dataset for Explainable Interpersonal Risk Factors of Mental Disturbance in Social Media Posts

Explainable Depression Symptom Detection in Social Media

CORec-Cri: How collaborative and social technologies can help to contextualize crises?

Rapid Classification of Crisis-Related Data on Social Networks using Convolutional Neural Networks

Reliability Analysis of Psychological Concept Extraction and Classification in User-penned Text

Threat or Opportunity? - Examining Social Bots in Social Media Crisis Communication

Causal Explanation Analysis on Social Media

Explainable AI for Mental Disorder Detection via Social Media: A survey and outlook

Leveraging Social Media to Predict COVID-19-Induced Disruptions to Mental Well-Being Among University Students: Modeling Study

Trustworthy Data and AI Environments for Clinical Prediction: Application to Crisis-Risk in People with Depression

Extracting Psychiatric Stressors for Suicide from Social Media Using Deep Learning.

Towards Understanding Emotions for Engaged Mental Health Conversations