Inpatient risk profile assignment

VALENF-based random forest classifier predicts inpatient risk profile assignment in humans (Macro-F1 0.825; 82.43% correct assignments).

“Random forest was selected because it combined adequate performance (macro-F1 0.825; 82.43% correct assignments) with the lowest net imbalance towards lower-risk assignments.”

Humansinternal evaluation across repeated train–test splitsn = 2,666

Nursing Assessment–Derived Risk Profiles and VALENF Instrument–Based Classification in Adult Inpatients: A Two-Stage Machine Learning Study

Research Square·6 Oct 2026Preprint