Contextualizing Datasets: Deepening Awareness of Data Biases through Critical Reflection of Civic Data

要旨

To broaden participation in computing and AI design, researchers urge supporting individuals to probe datasets for biases that technologies might amplify. Critically probing data requires first recognizing that it is not neutral, achievable through data contextualization---understanding data contexts including its origin and contents to recognize opportunities or shortcomings. We created a web-tool, “Contextualizing Datasets”, that helps users contextualize data by guiding them in data exploration and different stakeholder interactions. We applied this to an educational case study---using 311 data for allocating government flood resources---with a graduate class. Our findings suggest the tool helped scaffold students in contextualizing data to critically question data curation methods and stakeholder representation, with the local case study context supporting them to draw from lived experiences. From our results, we share ways to improve and re-purpose Contextualizing Datasets and reflect over how it can be leveraged to address generative AI concerns.

著者
Angie Zhang
The University of Texas at Austin, Austin, Texas, United States
Min Kyung Lee
University of Texas at Austin, Austin, Texas, United States

会議: CHI 2026

ACM CHI Conference on Human Factors in Computing Systems

セッション: Data Work

P1 - Room 131
7 件の発表
2026-04-15 20:15:00
2026-04-15 21:45:00