Mapping Violence Perceptions through YouTube Comments

A New Approach to Real-Time Violence Monitoring

Francesco Bailo
francesco.bailo@sydney.edu.au

Policy & Internet Conference 2026
University of Sydney

September 30, 2026

Published in EPJ Data Science

with

Ashani Amarasinghe\(^1\), Sascha Nanlohy\(^2\), Thomas Morgan\(^2\), David Hammond\(^2\), Yashdeep Dahiya\(^{1,3}\) and Francesco Bailo\(^1\)

DOI: 10.1140/epjds/s13688-026-00649-y

\(^1\)University of Sydney | \(^2\)Insititue for Economics and Peace | \(^3\)Monash University

Acknowledgement of Country

I would like to acknowledge the Traditional Owners of Australia and recognise their continuing connection to land, water and culture. The University of Sydney is located on the land of the Gadigal people of the Eora Nation. I pay my respects to their Elders, past and present.

Introduction

The Challenge of Measuring Violence

Traditional violence datasets face critical limitations:

  • Event-based datasets (ACLED, UCDP): Document actual violence through fatality counts
  • Cannot capture perceptions, fear, rumors, and community discourse
  • Marginalized and remote areas remain systematically underreported
  • Yet perceptions matter: Fear shapes behavior, economic activity, and social stability

Research Question

Can we systematically measure violence perceptions at scale using social media discourse?

  • Develop a Violence Perception Index (VPI) from geolocated YouTube comments
  • Validate against established violence indicators
  • Test whether VPI captures dynamics in underrepresented areas

The Violence Perception Index (VPI)

What Does the VPI Measure?

VPI quantifies intensity of violence-related discourse in geolocated comments:

  1. Direct experience: Eyewitness accounts
  2. Perceived threat: Fear and concern
  3. News circulation & rumors: Reported and unverified violence-related discourse
  4. Historical reference: Past violence patterns

Methodology

Data & Method Overview

Overview of data collection pipeline and methods

  • Scale: 1.2M videos | 14.8M comments (Jan 2020–June 2024) | 99.8% population coverage
  • Dictionary-based scoring: 118 weighted terms, validated against 4 LLMs (substantial agreement, Cohen’s κ 0.52–0.62)
  • Geographic aggregation: Inverse Distance Weighting → ~50km grid, monthly

Results

Strong Correlation with Realized Violence

Panel regression with comprehensive fixed effects:

Predictor Coefficient R²
ACLED Fatalities 0.0257*** 0.974
Homicides 0.0142*** 0.974
  • Grid, year, and month fixed effects
  • 37-67% increase in VPI per 1-SD increase in violence

The Key Finding: Geographic Heterogeneity

Split sample analysis (High vs. Low population grids):

ACLED (News-based):

  • High pop: Significant (0.0004***)
  • Low pop: Not significant

Official Homicides:

  • High pop: Not significant
  • Low pop: Significant (0.0004**)

Critical Insight

VPI correlates with ACLED in urban areas BUT with official records in marginalized areas where news coverage fails

Conclusion

Key Takeaways

  1. Violence perception is measurable at scale through social media discourse
  2. VPI correlates strongly with established violence indicators (R² = 0.97)
  3. Geographic heterogeneity is a feature: VPI captures discourse where traditional datasets fail — especially in marginalized/remote areas
  4. Immediately scalable across languages and geographies

Thank You

Questions?

Paper & Data:

Contact:

Francesco Bailo

University of Sydney

[francesco.bailo@sydney.edu.au]