>70%
machine learning model predicted cognitive symptoms from office environment data with reasonable accuracy
Exact: exceeds 70%
XGBoost model accuracy for predicting cognitive symptoms generally exceeded 70%
Using about four months of continuous indoor environmental quality (IEQ) data and ecological momentary assessments (EMAs) from ten office workers, XGBoost models were built per symptom and participant. Under random train-test splitting, accuracy generally exceeded 70%, though performance declined under temporal splitting, reflecting real-world challenges of predicting future states from past observations.
Under random train-test splitting, accuracy generally exceeded 70% with F1 scores ranging from 40% to 80%.
Related findings
higher indoor humidity tracks greater insomnia severity in older adults
Higher indoor humidity linked to greater insomnia severity in older adults
Youngmin Cho et al., 2025, Innovation in Aging
higher indoor CO2 levels linked to poorer delayed memory recall in older adults
Higher indoor CO2 levels associated with poorer delayed memory recall in older adults
Youngmin Cho et al., 2025, Innovation in Aging
evening indoor light closely tracks perceived stress in older adults
Higher evening indoor light exposure strongly linked to greater stress in older adults
Youngmin Cho et al., 2025, Innovation in Aging
Read more in
Workplace What your office costs
Four design variables that move cognitive performance and who pays for them
Education What school spaces do to children
Where the evidence on classroom air, acoustics, light and green is robust, where it is thin, and what to measure before the build.
Healthcare Built to Wake: How Hospital Noise and Light Undermine Patient Sleep
Of the two environmental levers on inpatient sleep, noise control is the better proven and the cheaper, while tunable lighting for the general ward is the one the evidence does not yet support.
More from The Built Review
Workplace Pricing biophilia: what the evidence is worth
Read at the primary sources, the business case for nature in buildings is narrower than advertised and strong enough to act on.
Workplace Germany's missing indoor-air bill
France, Britain and Australia have priced bad indoor air. Germany's missing number is a political choice, not a methodological limit.
Housing What insurers don't ask about buildings
Health insurers price age, tobacco and zip code. Building quality is in no model, and the law is only half the reason.