This project examined how children from groups underrepresented in STEM programs understand, interact with, and evaluate AI-driven digital assistants. 

This project examines the temporal dynamics of emotional appeals in Russian campaign messages used in the 2016 election on Facebook and Twitter. 

This project investigated how data ethics can be a point of departure in designing and evaluating good systems, examining the contradictions and pressure points among various data practices. 

This project addresses the conflict between convenience and privacy inherent to computer vision with the goal of developing future computer vision technologies that support diverse users with visual impairments, especially those who are traditionally underserved.   

The goal of this project is to build an AI system using crowd-sourced data to help predict the health impacts of different neighborhood environments.  

The team originally planned an extension of the UT1000 study to examine the relationship between biological indicators of stress, perception of stress, life events, and lifestyle in UT undergrads. This project addressed these challenges by developing and deploying environmental home beacons, wearable activity bands, and smartphone-based surveys.

In order to study the effectiveness of using a blend of sensing technology like smartphone apps and wearable technology, self-reported surveys, and readily available group-level information (grades, attendance), our team developed a study to understand more about the social lives, sleep habits, physical activity levels, and academic performance of 1000 UT undergraduate students in the 2019-2020 school year.

The Whole Communities–Whole Health team has built and tested a smart device that uses environmental sensors to better understand how indoor air quality affects family health by measuring indoor air pollution, carbon dioxide levels, humidity, and temperature in real-time.

Postpartum mood and anxiety disorders affect as many as one in five women globally. Left untreated, these can have major negative effects on maternal and child health and well-being. While effective treatments are available, nearly 60% of mothers with symptoms are undiagnosed, and 50% of diagnosed mothers are left untreated.

Artificial intelligence systems increasingly automate and assist people in making managerial and governance decisions. AI manages worker routines and tasks, determines how to distribute resources within cities, and assists in transportation management.