6th workshop on the

Data for the Wellbeing of the Most Vulnerable

At the the International AAAI Conference on Web and Social Media (ICWSM)

Location

Aalborg University

A. C. Meyers Vænge 15, 2450 København, Denmark

Building A, Room 2.1.083

Image credit: AP

The scale, reach, and real-time nature of the Internet is opening new frontiers for understanding the vulnerabilities in our societies, including inequalities and fragility in the face of a changing world. From tracking seasonal illnesses like the flu across countries and populations, to understanding the context of mental conditions such as anorexia and bulimia, web data has the potential to capture the struggles and wellbeing of diverse groups of people. Vulnerable populations including children, elderly, racial or ethnic minorities, socioeconomically disadvantaged, underinsured or those with certain medical conditions, are often absent in commonly used data sources. The recent developments in machine learning (e.g. large language models (LLMs)) render these issues even more urgent, introducing potential biases to these populations.

Thus, the aim of this workshop is to encourage the community to use new sources of data as well as methodologies to study the wellbeing of vulnerable populations. The selection of appropriate data sources, identification of vulnerable groups, and ethical considerations in the subsequent analysis are of great importance in the extension of the benefits of big data revolution to these populations. As such, the topic is highly multidisciplinary, bringing together researchers and practitioners in computer science, epidemiology, demography, linguistics, and many others.

We anticipate topics such as the below will be relevant:

Building Adaptive Cohorts: Identifying and Tracking Vulnerable Groups on Social Media
AI-Driven De-biasing: Mitigating Bias in Social Media Data for Equity and Inclusion
Using Individual-Level and Aggregate Data to Understand Vulnerability
Linking data to health and other well-being outcomes
Harnessing Diverse Population Data Sources for Robust Cross-Validation
Machine Learning Methods for Vulnerable Populations
Tracking Trends and Long-Term Vulnerabilities
Spatial, linguistic, and temporal analyses
Privacy, ethics, and informed consent in Social Media Research with Marginalized Communities
Biases and quality concerns around vulnerable groups in LLMs
Data quality and Algorithmic Fairness in research with marginalized communities

See previous workshop webpages: 2019, 2021, 2022, 2023, 2024

Keynote Speakers

Patrick Michael Brock

Senior Data Scientist, World Bank & UNHCR Joint Data Center

Roberta Sinatra

Professor in Social Data Science at the Center for Social Data Science, University of Copenhagen

Schedule

8:00 Introductions

8:05 Patrick Michael Brock: Innovating for better socioeconomic data on forced displacement

Global forced displacement is on the rise. To respond effectively, governments and their partners need data that is timely, high-quality, and accessible. But collecting this takes time and is expensive. Given the urgency and scale of the need, what can we do to accelerate and enhance the generation of socioeconomic data on forced displacement? One answer is to responsibly harness open data and innovative methods – carefully considering the specific ethical and legal dimensions of the protection that UNHCR provides, and the elimination of bias to ensure representativeness. To do this well, we need to work together, exchanging knowledge and tools between government, operations and research, so that we make progress as quickly as possible as a community of practice. I’ll discuss JDC work in this space, drawing out lessons, and linking to complementary ways in which UNHCR pursues its international protection mandate through combatting misinformation, disinformation and hate speech online.

8:40 Paper session I

Mapping Labor Market Vulnerability in the Age of AI: Evidence from Job Postings and Patent Data

Eun Cheol Choi, Qingyu Cao, Qi Guan, Shengzhu Peng, Po-Yuan Chen and Luca Luceri

Computationally Establishing Causality: The Impact of Demographic Minorities on Bias in Opinion Diffusion Models

Michalina Janik, Radosław Michalski and Akhil Arora

A Weak Supervision Learning Approach Towards an Equitable Mobility Estimation

Theophilus Aidoo, Till Koebe, Akansh Maurya, Hewan Shrestha and Ingmar Weber

Twitter anticipates adoption and change in pro-environmental behavior

Edoardo Maggioni, Luca Maria Aiello, Diego Garlaschelli and Rossana Mastrandrea

Coverage Biases in High-Resolution Satellite Imagery

Vadim Musienko, Axel Jacquet, Ingmar Weber and Till Koebe

9:50 Coffee break

10:10 Roberta Sinatra: When Algorithms Decide: Fairness, Bias, and Accountability in Decision Support Systems

Algorithms increasingly shape decisions that affect our lives, from college admissions to credit scoring to child protection. Yet, these systems often reproduce and amplify existing social biases or can even lead to discrimination. In this talk, I will unpack how algorithmic decision-making can go awry, drawing on a case study from Danish child welfare services, which used an algorithm to identify and assess children at risk from abuse. I will illustrate how even well-intentioned quantitative systems can encode discrimination and obscure accountability. In the talk I will also discuss methodological innovations to audit algorithms and correct unfair algorithmic outputs. By bridging computer science and social science, this talk aims to offer both critical insights and practical pathways toward more equitable algorithmic systems.

10:45 Paper session II

A Multimodal TikTok Dataset of Ecuador's 2024 Political Crisis and Organized Crime Discourse

Gabriela Pinto and Emilio Ferrara

Targeting and Messaging in Political Ads on Meta platforms During Brazil's 2024 Municipal Elections

Daniel Verdi Do Amarante and Francesco Pierri

Gender Bias in How Public Figures are Referenced in News Quotes at Scale

Justyna Janczy, Marko Čuljak, Andreas Spitz and Akhil Arora

EDTok: A Dataset for Eating Disorder Content on TikTok

Charles Bickham, Bryan Ramirez-Gonzalez, Minh Duc Chu, Kristina Lerman and Emilio Ferrara

“This actually goes beyond just pronouns”: Shifting discourses on trans-affirming features in tech

Cedar Brown

12:00 End

Call for papers & abstracts

The call is closed. See accepted papers in the ICWSM Workshop Proceedings.

We welcome both 2-page abstracts, as well as Long (8 pages) and Short (4 pages) papers - not counting references & appendix.

In the case of abstracts, the 2-page limit applies to text only.

All submissions should be in English. The Long and Short papers will be published in ICWSM Workshop proceedings by the AAAI Press. Please follow the AAAI format.

The reviewing process will be double blind, so please anonymize your submissions.

Submit via the EasyChair portal: https://easychair.org/conferences/?conf=dwmv25 (closed)

Program Committee: