The Problem of Semantic Shift in Longitudinal Monitoring of Social Media: A Case Study on Mental Health During the COVID-19 Pandemic

TL;DR

This study investigates how semantic shifts in social media language during COVID-19 affect mental health analysis, revealing that unstable features can distort results and that measuring semantic shift can enhance model robustness.

Contribution

It demonstrates the impact of semantic shift on longitudinal mental health analysis and proposes using semantic shift measurement to improve model robustness.

Findings

Semantic shift can cause significant changes in longitudinal estimates.

A new method can identify failure points in language models.

Measuring semantic shift improves predictive robustness.

Abstract

Social media allows researchers to track societal and cultural changes over time based on language analysis tools. Many of these tools rely on statistical algorithms which need to be tuned to specific types of language. Recent studies have shown the absence of appropriate tuning, specifically in the presence of semantic shift, can hinder robustness of the underlying methods. However, little is known about the practical effect this sensitivity may have on downstream longitudinal analyses. We explore this gap in the literature through a timely case study: understanding shifts in depression during the course of the COVID-19 pandemic. We find that inclusion of only a small number of semantically-unstable features can promote significant changes in longitudinal estimates of our target outcome. At the same time, we demonstrate that a recently-introduced method for measuring semantic shift may be used to proactively identify failure points of language-based models and, in turn, improve predictive generalization.