Automatically Estimating the Severity of Multiple Symptoms Associated with Depression

This chapter describes the participation of the RELAI team in the eRisk 2020 second task. The 2020 edition of eRisk proposed two tasks: (T1) Early assessment of the risk of self-harm and (T2) Signs of depression in social media users. The second task focused on automatically filling a depression questionnaire given the user’s writing history. The RELAI team addressed it using Latent Dirichlet Allocation (LDA), a topic modelling algorithm, and an approach based on writing styles. The proposed system based on LDA performed well according to all the evaluation metrics. According to the Average Difference between Overall Depression Levels (ADODL), it achieved the best performance among participants with a score of 83.15%. Overall, the submitted systems achieved promising results and suggested that social media evidence could help early mental health risk assessment.