US 12,443,861 B2
Multi-station decision network
Christopher Worsley, Alpine, UT (US); Ryan William Johnson, Alpine, UT (US); Jeffery Theodore Stockett, Lehi, UT (US); David Roy Watson, Jr., Pleasant Grove, UT (US); Austin Paul Bagley, Mapleton, UT (US); and Logan Brent Uibel, Vineyard, UT (US)
Assigned to Rithm Software Inc., Lindon, UT (US)
Filed by Rithm Software Inc., Lindon, UT (US)
Filed on Sep. 13, 2022, as Appl. No. 17/931,736.
Prior Publication US 2024/0086742 A1, Mar. 14, 2024
Int. Cl. G06N 20/00 (2019.01); G06F 9/50 (2006.01); G06N 5/04 (2023.01)
CPC G06N 5/04 (2013.01) [G06F 9/5077 (2013.01); G06N 20/00 (2019.01)] 18 Claims
OG exemplary drawing
 
1. A computer system comprising:
memory; and
one or more processors in communication with the memory and configured with processor-executable instructions to perform operations comprising:
configuring a decision network, wherein configuring the decision network comprises:
generating a user interface comprising selectable options for configuring a plurality of stations within the decision network;
based at least in part on user selections made within the user interface, generating and storing a first station configuration for a first station in the decision network, wherein the first station configuration identifies each of at least: one or more data fields associated with a container object, access restrictions associated with the first station, at least one data acquisition source for the first station, and executable logic for generating a first subset of field values with respect to the container object;
generating and storing a second station configuration for a second station in the decision network, wherein the second station configuration includes executable logic for generating a second subset of field values with respect to the container object;
receiving, via user interaction with at least one user interface, a request to generate (a) a first virtual connection between the first station and the second station and (b) a second virtual connection between the second station and a third station, wherein the first and second virtual connections indicate to a computing system that instances of the container object should virtually move from the first station to the second station to the third station during operation of the decision network; and
storing indications of (a) the first virtual connection in association with the first station configuration and (b) the second virtual connection in association with the second station configuration; and
executing an instance of the decision network, wherein executing the instance of the decision network comprises:
applying the executable logic of the first station with respect to data from the at least one data acquisition source for the first station to generate and store the first subset of field values within a first instance of the container object;
subsequent to modification of the first instance of the container object by the first station, applying the executable logic of the second station with respect to data from the at least one data acquisition source for the second station to generate and store the second subset of field values within the first instance of the container object, wherein the second subset of field values include at least one value representing a prediction or probability determined based on data accessible to the second station; and
applying executable logic of the third station to generate a final decision of the instance of the decision network based on the data stored in the first instance of the container object by each of the first and second stations, wherein the prediction or probability determined at the second station contributes to the final decision generated at the third station.