US 11,776,008 B2
Contextual prediction system and method
Bruce Eaton Buchanan, Brooklyn, NY (US); Stuart James FitzRoy, Artarmon (AU); Scott Matthew Jackson, Baulkham Hills (AU); Kanishka Mohaia, Eastwood (AU); Dinesh Gurram, Baulkham Hills (AU); and Yan Xu, Wentworth (AU)
Assigned to ROKT PTE. LTD., Singapore (SG)
Filed by Rokt Pte. Ltd., Singapore (SG)
Filed on Mar. 18, 2020, as Appl. No. 16/822,730.
Prior Publication US 2021/0295382 A1, Sep. 23, 2021
Int. Cl. G06Q 30/00 (2023.01); G06Q 30/0251 (2023.01); G06F 16/9535 (2019.01); G06N 5/04 (2023.01); G06N 20/00 (2019.01)
CPC G06Q 30/0253 (2013.01) [G06F 16/9535 (2019.01); G06N 5/04 (2013.01); G06N 20/00 (2019.01); G06Q 30/0269 (2013.01)] 18 Claims
OG exemplary drawing
 
11. An online content provisioning system for use in relation to online electronic transactions, said content provisioning system comprising:
a computer-implemented server for conducting an electronic transaction with a user via a website, wherein each of said server and said website are coupled to a communications network, said website displaying a webpage encoded in accordance with a content provisioning application programming interface (API) and including a contextual content display region, said webpage being further encoded with a link to a content provisioning software development kit (SDK) associated with said content provisioning platform;
an inventory provider server coupled to said communications network; and
a computer-implemented content provisioning platform coupled to said communications network, said content provisioning platform including:
a selection engine that includes a prediction services module implemented using a machine learning algorithm and configured to:
in response to a request from said webpage during a checkout process of said electronic transaction, said request being sent in response to said link being activated during checkout process, determine a likelihood of the user to engage with respective inventory based on a contextual attribute associated with said electronic transaction and a customer attribute associated with a user performing said electronic transaction; and
select inventory, based on said likelihood, from available inventory derived from said inventory provider server during the electronic transaction; and
a content rendering module that:
stores data that determines how content is to be rendered on said e-commerce webpage, said data being selected from the group consisting of campaign meta data, creatives, UX formats, and UX designs; and
formats said selected inventory for display on said webpage, based on said stored data; and
wherein said content provisioning API is associated with said content provisioning platform and enables interaction among said website, said content provisioning platform, and said inventory provider server, and
further wherein said content provisioning API delivering said selected content for display in said contextual content display region of said webpage.