US 11,790,406 B2
Systems and methods for improved online predictions
Xuan Cao, Fremont, CA (US); Georgios Rovatsos, San Francisco, CA (US); and Wei Shen, Pleasanton, CA (US)
Assigned to WALMART APOLLO, LLC, Bentonville, AR (US)
Filed by Walmart Apollo, LLC, Bentonville, AR (US)
Filed on Jan. 31, 2022, as Appl. No. 17/588,778.
Prior Publication US 2023/0245177 A1, Aug. 3, 2023
Int. Cl. G06Q 30/0273 (2023.01); G06Q 30/0241 (2023.01); G06Q 30/0242 (2023.01)
CPC G06Q 30/0275 (2013.01) [G06Q 30/0242 (2013.01); G06Q 30/0249 (2013.01); G06Q 30/0276 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A system comprising:
one or more processors; and
one or more non-transitory computer-readable media storing computing instructions that, when executed on the one or more processors cause the one or more processors to perform operations comprising:
(1) receiving a request to generate one or more campaigns;
(2) determining, by a predictive algorithm via a machine learning model, one or more predicted bids for one or more keywords in the one or more campaigns, wherein the predictive algorithm uses historical data of historical campaigns as a training data set and outputs the one or more predicted bids for the one or more keywords in the one or more campaigns;
(3) adjusting the one or more predicted bids for the one or more keywords in the one or more campaigns by multiplying the one or more predicted bids by a square root of a quotient of a return on advertisement spend (ROAS) for an item divided by a ROAS received with the request to generate the one or more campaigns;
(4) pacing the one or more predicted bids, as adjusted, for the one or more keywords in the one or more campaigns, wherein pacing the one or more predicted bids, as adjusted, comprises multiplying the one or more predicted bids, as adjusted, by a pacing factor; and
iterating (2)-(4) at one or more periodic intervals as real-time data from the one or more campaigns is added to the training data set, wherein the iterating comprises repeated cycling of (2)-(4) to prevent over-predictions of predictive bids, and wherein the real-time data of performance of the one or more campaigns is iteratively added to the training data set for the predictive algorithm, as trained.