Evaluating Lexicon Incorporation for Depression Symptom Estimation

TL;DR

This study investigates how integrating sentiment, emotion, and domain-specific lexicons into transformer models affects depression symptom estimation, achieving state-of-the-art results by enhancing prediction accuracy with external knowledge.

Contribution

It introduces a method for incorporating various lexicons into transformer models and demonstrates improved depression level estimation performance, setting new benchmarks.

Findings

Lexicon integration improves prediction accuracy.

Different lexicons exhibit distinct effects depending on the task.

Achieved state-of-the-art results in depression level estimation.

Abstract

This paper explores the impact of incorporating sentiment, emotion, and domain-specific lexicons into a transformer-based model for depression symptom estimation. Lexicon information is added by marking the words in the input transcripts of patient-therapist conversations as well as in social media posts. Overall results show that the introduction of external knowledge within pre-trained language models can be beneficial for prediction performance, while different lexicons show distinct behaviours depending on the targeted task. Additionally, new state-of-the-art results are obtained for the estimation of depression level over patient-therapist interviews.