Evaluating Lexicon Incorporation for Depression Symptom Estimation

Kirill Milintsevich, Gaël Dias, Kairit Sirts
4/30/2024
cs.CLcs.AI

No official code found. Best implementation: 501Good/dialogue-classifier (1 star, MIT, last commit 2024-07-04).

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.

View on arXivView PDF

Code Implementations(1)

501Good/dialogue-classifierCommunity100%
MIT
10PythonFeb 1, 20242 years agoMIT

Similar Papers

Depression and Anxiety Prediction Using Deep Language Models and Transfer Learning

86% match2024

DepressLLM: Interpretable domain-adapted language model for depression detection from real-world narratives

85% match2025

Enhancing Depression Detection with Chain-of-Thought Prompting: From Emotion to Reasoning Using Large Language Models

85% match2025

Depression Detection on Social Media with Large Language Models

84% match2024

AI in Mental Health: Emotional and Sentiment Analysis of Large Language Models' Responses to Depression, Anxiety, and Stress Queries

84% match2025

Leveraging Large Language Models for Cost-Effective, Multilingual Depression Detection and Severity Assessment

83% match2025

Cite this paper

@article{milintsevich2024evaluating,
  title  = {Evaluating Lexicon Incorporation for Depression Symptom Estimation},
  author = {Kirill Milintsevich and Gaël Dias and Kairit Sirts},
  year   = {2024},
  eprint = {2404.19359},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2404.19359v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2404.19359.svg)](https://codeofpaper.com/paper/2404.19359)

Discussion