US 11,734,606 B2
Method for predicting operational effectiveness
Dennis Bushmitch, Somerset, NJ (US); and Richard Cozby, Bel Air, MD (US)
Assigned to The Government of the United States as represented by the Secretary of the Army, Washington, DC (US)
Filed by The Government of the United States, as represented by the Secretary of the Army, Washington, DC (US)
Filed on Jun. 25, 2020, as Appl. No. 16/911,443.
Application 16/911,443 is a division of application No. 14/499,297, filed on Sep. 29, 2014, granted, now 10,733,525.
Prior Publication US 2020/0327452 A1, Oct. 15, 2020
Int. Cl. G06N 20/00 (2019.01)
CPC G06N 20/00 (2019.01) 2 Claims
OG exemplary drawing
 
1. A method, comprising:
predicting a first operational score for a first individual service by way of a first service adaptive learning system;
predicting a second operational score for a second individual service by way of a second service adaptive learning system that is distinct from the first service adaptive learning system; and
predicting an overall operational effectiveness through use of an operational adaptive learning system,
where the overall operational effectiveness is based, at least in part, on the first operational score and the second operational score,
where the first service adaptive learning system is trained from a first set of training technical performance data and a first set of training operational assessment data,
where the first service adaptive learning system is trained from a second set of training technical performance data and a second set of training operational assessment data,
where the first individual service and the second individual service are on the same hierarchical level,
where the first service adaptive learning system is trained prior to predicting the first operational score,
where the first service adaptive learning system is validated after being trained and prior to predicting the first operational score,
where the second service adaptive learning system is trained prior to predicting the second operational score,
where the second service adaptive learning system is validated after being trained and prior to predicting the second operational score,
where the first service adaptive learning system is trained while online,
where the first service adaptive learning system is validated while online,
where the first service adaptive learning system is a first service adaptive deep learning system of at least five layers,
where the second service adaptive learning system is trained while online,
where the second service adaptive learning system is validated while online,
where the second service adaptive learning system is a second service adaptive deep learning system of at least five layers,
where the first set of training operational assessment data is human produced,
where the first set of training technical performance data and the second set of training technical performance data is identical, and
where the first set of training operational assessment data and the second set of training technical performance data is identical.