Incremental Value of Smartphone Sensing for Monitoring Momentary Affect Intensity in Adults Using Transformer-Based Models: Observational Study
Yiqin Zhu, Yuyi Yang, Renee J Thompson, Lorraine Buis · 2026
Read Incremental Value of Smartphone Sensing for Monitoring Momentary Affect Intensity in Adults Using Transformer-Based Models: Observational Study by Yiqin Zhu, Yuyi Yang, Renee J Thompson, Lorraine Buis (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13626072.txt.
Description
Yiqin Zhu, Yuyi Yang, Renee J Thompson, and Lorraine Buis ask how much a phone's sensors add to forecasting a person's mood, moment by moment. One hundred thirty-three adults spent fourteen days answering five short mood surveys a day while their phones quietly logged sensing data. The team trained transformer models across windows from one hour to two days. One design merged personal history, passive sensing, and past moods. The models also learned where people said they were, such as work.
Personal averages alone already forecast mood fairly well. The full sensing models did better for both low and high mood, lifting accuracy by a few points over those baselines. Splitting gains by data source and time window shows designers where the predictive juice sits, guiding leaner mood-tracking tools.
Catalog record
| CREATOR | Yiqin Zhu, Yuyi Yang, Renee J Thompson, Lorraine Buis |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 63 KB |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |