An End-to-End Set Transformer for User-Level Classification of Depression and Gambling Disorder

TL;DR

This paper introduces an end-to-end set transformer architecture for user-level classification of depression and gambling disorder using social media posts, improving interpretability and performance over previous post-level methods.

Contribution

The work presents a novel permutation-invariant transformer model that processes sets of user posts without positional encodings, enabling better detection and interpretability in mental health classification.

Findings

Achieved the best ERDE5 score of 0.015 for gambling detection.

Obtained second-best ERDE50 score of 0.009 for gambling detection.

Secured second-best ERDE50 score of 0.027 for depression detection.

Abstract

This work proposes a transformer architecture for user-level classification of gambling addiction and depression that is trainable end-to-end. As opposed to other methods that operate at the post level, we process a set of social media posts from a particular individual, to make use of the interactions between posts and eliminate label noise at the post level. We exploit the fact that, by not injecting positional encodings, multi-head attention is permutation invariant and we process randomly sampled sets of texts from a user after being encoded with a modern pretrained sentence encoder (RoBERTa / MiniLM). Moreover, our architecture is interpretable with modern feature attribution methods and allows for automatic dataset creation by identifying discriminating posts in a user's text-set. We perform ablation studies on hyper-parameters and evaluate our method for the eRisk 2022 Lab on early detection of signs of pathological gambling and early risk detection of depression. The method proposed by our team BLUE obtained the best ERDE5 score of 0.015, and the second-best ERDE50 score of 0.009 for pathological gambling detection. For the early detection of depression, we obtained the second-best ERDE50 of 0.027.