US 12,260,314 B2
Machine-based prediction of visitation caused by viewing
Shawn Spooner, Beaverton, OR (US)
Assigned to Billups LLC, Lake Oswego, OR (US)
Filed by BILLUPS, INC., Lake Oswego, OR (US)
Filed on May 22, 2023, as Appl. No. 18/200,093.
Application 18/200,093 is a continuation of application No. 16/170,955, filed on Oct. 25, 2018, granted, now 11,694,063.
Prior Publication US 2023/0297813 A1, Sep. 21, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06N 3/045 (2023.01); G06F 16/901 (2019.01); G06F 16/907 (2019.01); G06F 17/18 (2006.01); H04W 4/02 (2018.01)
CPC G06N 3/045 (2023.01) [G06F 16/9024 (2019.01); G06F 16/907 (2019.01); G06F 17/18 (2013.01); H04W 4/023 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method for machine-based prediction of visitation, the method comprising:
obtaining first trace data including previously traversed routes for multiple devices, the first trace data having missing data for unobserved portions of the previously traversed routes, each of the previously traversed routes that includes an unobserved portion including at least an observed portion within the first trace data;
generating, by a first machine-learned network trained using deep learning, second trace data that fills in at least some of the missing data for the unobserved portions of the previously traversed routes, the second trace data indicating that first content was perceptible from at least one of the unobserved portions of the previously traversed routes;
after generating the second trace data and using the second trace data as an input, determining exposure to the first content;
modeling, by a second machine-learned network and based on the determination of exposure to the first content, a causal effect of the first content to visits of a location; and
displaying a prediction of effectiveness of second content based on the causal effect.