US 11,941,661 B2
Method, apparatus, and computer program product for predicting web browsing behaviors of consumers
Michael Sussman, Mountain View, CA (US); Jesse Pinho, Berlin (DE); Michael Hines, Chicago, IL (US); Jim Challenger, Chicago, IL (US); David Hanley, Salt Lake City, UT (US); Isaac Sanders, Chicago, IL (US); and Dean Marano, Grand Rapids, MI (US)
Assigned to Groupon, Inc., Chicago, IL (US)
Filed by Groupon, Inc., Chicago, IL (US)
Filed on Dec. 21, 2021, as Appl. No. 17/557,763.
Application 17/557,763 is a continuation of application No. 16/845,617, filed on Apr. 10, 2020, granted, now 11,238,496.
Application 16/845,617 is a continuation of application No. 15/280,738, filed on Sep. 29, 2016, granted, now 10,740,793, issued on Aug. 11, 2020.
Claims priority of provisional application 62/235,142, filed on Sep. 30, 2015.
Prior Publication US 2022/0222705 A1, Jul. 14, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. G06Q 30/02 (2023.01); G06Q 30/0241 (2023.01); G06Q 30/0251 (2023.01); G06F 16/835 (2019.01)
CPC G06Q 30/0255 (2013.01) [G06Q 30/0277 (2013.01); G06F 16/835 (2019.01)] 20 Claims
OG exemplary drawing
 
1. An apparatus comprising at least one processor and at least one non-transitory memory comprising program code, wherein the at least one non-transitory memory and the program code are configured to, with the at least one processor, cause the apparatus to:
generate a link graph for a website, the website comprising a plurality of webpages, the link graph generated based at least in part on data comprising one or more hyperlinks and historical data associated with the plurality of webpages;
wherein the link graph comprises a plurality of nodes and a plurality of links,
wherein a node of the plurality of nodes represents a webpage of the plurality of webpages and a link of the plurality of links represents a hyperlink of the one or more hyperlinks, and wherein each link is associated with a weight value;
generate a threshold for each weight value based at least in part on one or more of a desired computational expense or a desired accuracy;
eliminate those links of the plurality of links from the link graph that are associated with a weight value below the threshold; and
generate a programmatically expected path for a particular user through the link graph based at least in part on one or more weight values associated with a first node.