UQ at #SMM4H 2023: ALEX for Public Health Analysis with Social Media

TL;DR

This paper introduces the ALEX framework that leverages large language models with explanation mechanisms to enhance public health analysis from social media data, addressing data imbalance and cost issues.

Contribution

The paper presents a novel ALEX framework that improves social media-based public health analysis by utilizing LLM explanations and data balancing techniques.

Findings

ALEX achieved top performance in SMM4H 2023 tasks.

Effective handling of data imbalance through augmentation.

Utilization of LLM explanations improved analysis accuracy.

Abstract

As social media becomes increasingly popular, more and more activities related to public health emerge. Current techniques for public health analysis involve popular models such as BERT and large language models (LLMs). However, the costs of training in-domain LLMs for public health are especially expensive. Furthermore, such kinds of in-domain datasets from social media are generally imbalanced. To tackle these challenges, the data imbalance issue can be overcome by data augmentation and balanced training. Moreover, the ability of the LLMs can be effectively utilized by prompting the model properly. In this paper, a novel ALEX framework is proposed to improve the performance of public health analysis on social media by adopting an LLMs explanation mechanism. Results show that our ALEX model got the best performance among all submissions in both Task 2 and Task 4 with a high score in Task 1 in Social Media Mining for Health 2023 (SMM4H)[1]. Our code has been released at https:// github.com/YanJiangJerry/ALEX.