Time-based splitting strategy and forecast horizons.

a. The test set contained the latest 7 surveys from each participant, the validation set contained the previous 7 surveys, and the training set included all (>7) earlier surveys b. Each label to predict was paired with 3 days of input data (12 6-hour...

Full description

Saved in:

Bibliographic Details
Main Author:	Thierry Jean (20691795) (author)
Other Authors:	Rose Guay Hottin (20691798) (author), Pierre Orban (62056) (author)
Published:	2025
Subjects:	Neuroscience Science Policy Mental Health Biological Sciences not elsewhere classified Information Systems not elsewhere classified thus providing richer results also showed previous research focused prediction targets derived passive sensor data mean absolute error machine learning algorithm g ., calm forecasting mental states extreme gradient boosting digital phenotyping studies clinical rating scales forecast mental states forecast horizon expanded 3 forecast horizons systematically overestimated performance often ordinal nature next week ). evaluating predictive performance 2 predictive tasks 2 learning algorithms xgboost models performed using performance metrics forecast tasks ordinal regression xlink "> term memory small decrease similar levels significant yet seeing things remains unclear popular choices next day lstm models long short continuous regression binary classification best suited 73 %, 551 days
Tags:	Add Tag No Tags, Be the first to tag this record!

Time-based splitting strategy and forecast horizons.

Similar Items