Using text classification to estimate the depression level of reddit users

Abstract

Psychologists have used tests and carefully designed survey questions, such as Beck’s Depression Inventory (BDI), to identify the presence of depression and to assess its severity level. On the other hand, methods for automatic depression detection have gained increasing interest since all the information available in social media, such as Twitter and Facebook, enables novel approaches based on language use. More precisely, these methods have focused on learning to detect depressive users through their language usage. However, little effort has been put into going beyond mere detection, towards estimating users’ actual clinical depression level. The present study is a first step towards that direction: we try to develop a model able to estimate Reddit’s users’ clinical depression level by filling in the BDI depression questionnaire on behalf of each user. To carry out his task, the model answers all 21 questions of the questionnaire using the confidence value outputted by a binary text classifier trained to detect depressed users on Reddit. Our proposal was publicly tested in the CLEF’s eRisk 2019 lab obtaining the best and second-best performance among the other 13 submitted models.