Multifeature fusion attention network for suicide risk assessment based on social media: Algorithm development and validation

Abstract

Background: Suicide has become the fifth leading cause of death worldwide. With development of the internet, social media has become an imperative source for studying psychological illnesses such as depression and suicide. Many methods have been proposed for suicide risk assessment. However, most of the existing methods cannot grasp the key information of the text. To solve this problem, we propose an efficient method to extract the core information from social media posts for suicide risk assessment. Objective: We developed a multifeature fusion recurrent attention model for suicide risk assessment. Methods: We used the bidirectional long short-term memory network to create the text representation with context information from social media posts. We further introduced a self-attention mechanism to extract the core information. We then fused linguistic features to improve our model. Results: We evaluated our model on the dataset delivered by the Computational Linguistics and Clinical Psychology 2019 shared task. The experimental results showed that our model improves the risk-F1, urgent-F1, and existence-F1 by 3.3%, 0.9%, and 3.7%, respectively. Conclusions: We found that bidirectional long short-term memory performs well for long text representation, and the attention mechanism can identify the key information in the text. The external features can complete the semantic information lost by the neural network during feature extraction and further improve the performance of the model. The experimental results showed that our model performs better than the state-of-the-art method. Our work has theoretical and practical value for suicidal risk assessment.