Granted Patent B2

US 12,499,470 · App. 17/527,595 · Granted Dec 16, 2025

Inventors: Charith Srian Peris (Cambridge, MA); Saket Subhash Mengle (Mansfield, MA); Beth Teresa Logan (Cambridge, MA); Willard Lennox Simmons (Boston, MA)

Assignee: Roku DX Holdings, Inc.

G06Q30/0275G06N5/04G06N20/00G06Q30/0243G06Q30/0244G06Q30/0246G06Q30/0277G06N3/02

Abstract

Claims (60)

1 . A method comprising:

an advertisement exchange server for receiving, via a network connection, bid requests from publishers having advertisement inventory, each bid request being directed to a designated advertisement type, the advertisement exchange server being configured for:

determining website history information about a plurality of consumers, the website history information comprising one or more websites visited by each of the plurality of consumers prior to a display of an advertisement;

determining ad history information about the plurality of consumers, the ad history information comprising one or more ads that were provided to each of the plurality of consumers prior to the display of the advertisement;

providing the advertisement for display on a computing device, for each consumer of at least a subset of consumers of the plurality of consumers, responsive to a first bid opportunity for the advertisement to be displayed;

tracking website actions of the subset of consumers to whom the advertisement was provided for display;

generating, by an artificial intelligence module, a predictive model that the bid opportunity lead to a conversion based on the website history information, the ad history information, and the tracking of website actions after the advertisement was provided for display;

generating a price to bid for a second bid opportunity, to display the advertisement, based on a prediction generated based on the predictive model;

submitting the generated price to bid for the second bid opportunity to display the advertisement to a marketplace, the marketplace comprising one or more client computing devices for generating media rich advertisements having embedded metadata associated therewith, the embedded metadata including one or more classifications for categorizing each generated media rich advertisement as to type, each one or more client computing devices including a communications module for transmitting respective generated media rich advertisements via the network connection, wherein the marketplace either accepts or rejects the generated price for the second bid opportunity;

receiving an indication that the generated price was accepted;

receiving tracking data comprising website actions of one or more consumers to whom the advertisement was provided for display in accordance with the second bid opportunity, wherein the tracking data indicates whether the second bid opportunity led to an impression or conversion; and

providing the prediction and the tracking data to the artificial intelligence module which is configured to calibrate and error correct the predictive model to generate a refined model that is applied to make a better prediction based on the tracking data.

2 . The method of claim 1 , wherein the ad history information comprises interactions with the one or more ads by the plurality of consumers.

3 . The method of claim 1 , wherein the conversion comprises an interaction with the advertisement.

4 . The method of claim 1 , wherein the conversion comprises a purchase of a product associated with the advertisement.

5 . The method of claim 4 , wherein the predictive model indicates which consumers would have purchased the product without the advertisement.

6 . The method of claim 1 , further comprising:

providing the advertisement for display in accordance with the second bid opportunity.

7 . The method of claim 1 , further comprising:

refining, by the artificial intelligence module, the predictive model based on the tracking of website actions of one or more consumers to whom the advertisement was provided for display in accordance with the second bid opportunity.

8 . A system, comprising:

an advertisement exchange server for receiving, via a network connection, bid requests from publishers having advertisement inventory, each bid request being directed to a designated advertisement type, the advertisement exchange server being configured for:

determining website history information about a plurality of consumers, the website history information comprising one or more websites visited by each of the plurality of consumers prior to a display of an advertisement;

determining ad history information about the plurality of consumers, the ad history information comprising one or more ads that were provided to each of the plurality of consumers prior to the display of the advertisement;

providing the advertisement for display on a computing device, for each consumer of at least a subset of consumers of the plurality of consumers, responsive to a first bid opportunity for the advertisement to be displayed;

tracking website actions of the subset of consumers to whom the advertisement was provided for display;

generating, by an artificial intelligence module, a predictive model that the bid opportunity lead to a conversion based on the website history information, the ad history information, and the tracking of website actions after the advertisement was provided for display;

generating a price to bid for a second bid opportunity, to display the advertisement, based on a prediction generated based on the predictive model;

submitting the generated price to bid for the second bid opportunity to display the advertisement to a marketplace, the marketplace comprising one or more client computing devices for generating media rich advertisements having embedded metadata associated therewith, the embedded metadata including one or more classifications for categorizing each generated media rich advertisement as to type, each one or more client computing devices including a communications module for transmitting respective generated media rich advertisements via the network connection, wherein the marketplace either accepts or rejects the generated price for the second bid opportunity;

receiving an indication that the generated price was accepted;

receiving tracking data comprising website actions of one or more consumers to whom the advertisement was provided for display in accordance with the second bid opportunity, wherein the tracking data indicates whether the second bid opportunity led to an impression or conversion; and

providing the prediction and the tracking data to the artificial intelligence module which is configured to calibrate and error correct the predictive model to generate a refined model that is applied to make a better prediction based on the tracking data.

9 . The system of claim 8 , wherein the ad history information comprises interactions with the one or more ads by the plurality of consumers.

10 . The system of claim 8 , wherein the conversion comprises an interaction with the advertisement.

11 . The system of claim 8 , wherein the conversion comprises a purchase of a product associated with the advertisement.

12 . The system of claim 11 , wherein the predictive model indicates which consumers would have purchased the product without the advertisement.

13 . The system of claim 8 , the advertisement exchange server being further configured for:

providing the advertisement for display in accordance with the second bid opportunity.

14 . The system of claim 13 , the advertisement exchange server being further configured for:

refining, by the artificial intelligence module, the predictive model based on the tracking of website actions of one or more consumers to whom the advertisement was provided for display in accordance with the second bid opportunity.

15 . A non-transitory processor-readable medium having one or more instructions operational on an advertisement exchange server for receiving, via a network connection, bid requests from publishers having advertisement inventory, each bid request being directed to a designated advertisement type, which, when executed by a processor, cause the advertisement exchange server to perform operations comprising:

determining website history information about a plurality of consumers, the website history information comprising one or more websites visited by each of the plurality of consumers prior to a display of an advertisement;

determining ad history information about the plurality of consumers, the ad history information comprising one or more ads that were provided to each of the plurality of consumers prior to the display of the advertisement;

providing the advertisement for display on a computing device, for each consumer of at least a subset of consumers of the plurality of consumers, responsive to a first bid opportunity for the advertisement to be displayed;

tracking website actions of the subset of consumers to whom the advertisement was provided for display;

generating, by an artificial intelligence module, a predictive model that the bid opportunity lead to a conversion based on the website history information, the ad history information, and the tracking of website actions after the advertisement was provided for display;

generating a price to bid for a second bid opportunity, to display the advertisement, based on a prediction generated based on the predictive model;

submitting the generated price to bid for the second bid opportunity to display the advertisement to a marketplace, the marketplace comprising one or more client computing devices for generating media rich advertisements having embedded metadata associated therewith, the embedded metadata including one or more classifications for categorizing each generated media rich advertisement as to type, each one or more client computing devices including a communications module for transmitting respective generated media rich advertisements via the network connection, wherein the marketplace either accepts or rejects the generated price for the second bid opportunity;

receiving an indication that the generated price was accepted;

receiving tracking data comprising website actions of one or more consumers to whom the advertisement was provided for display in accordance with the second bid opportunity, wherein the tracking data indicates whether the second bid opportunity led to an impression or conversion; and

providing the prediction and the tracking data to the artificial intelligence module which is configured to calibrate and error correct the predictive model to generate a refined model that is applied to make a better prediction based on the tracking data.

16 . The non-transitory processor-readable medium of claim 15 , wherein the ad history information comprises interactions with the one or more ads by the plurality of consumers.

17 . The non-transitory processor-readable medium of claim 15 , wherein the conversion comprises an interaction with the advertisement.

18 . The non-transitory processor-readable medium of claim 15 , wherein the conversion comprises a purchase of a product associated with the advertisement.

19 . The non-transitory processor-readable medium of claim 18 , wherein the predictive model indicates which consumers would have purchased the product without the advertisement.

20 . The non-transitory processor-readable medium of claim 15 , the operations further comprising:

providing the advertisement for display in accordance with the second bid opportunity; and

refining, by the artificial intelligence module, the predictive model based on the tracking of website actions of one or more consumers to whom the advertisement was provided for display in accordance with the second bid opportunity.

21 . The method of claim 1 , wherein the tracking is performed across a plurality of bid opportunities, impressions, and conversions.

22 . The method of claim 1 , wherein the bid requests are separated into both a treatment group and a control group.

Assignments (1)

Continuity (3)

References Cited (44)