US 11,861,664 B2
Keyword bids determined from sparse data
Anirban Basu, Bengaluru (IN); Tathagata Sengupta, Bangalore (IN); Kunal Kumar Jain, Chennai (IN); and Ashish Kumar, Bangalore (IN)
Assigned to Adobe Inc., San Jose, CA (US)
Filed by Adobe Inc., San Jose, CA (US)
Filed on Sep. 29, 2022, as Appl. No. 17/955,781.
Application 17/955,781 is a continuation of application No. 16/555,380, filed on Aug. 29, 2019, granted, now 11,494,810.
Prior Publication US 2023/0021653 A1, Jan. 26, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06Q 30/00 (2023.01); G06Q 30/0273 (2023.01); G06F 16/9038 (2019.01)
CPC G06Q 30/0275 (2013.01) [G06F 16/9038 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
receiving, by a processing device, a portfolio of keywords including low-impression keywords from multiple client devices;
identifying, by the processing device, the low-impression keywords by determining which keywords of the portfolio of keywords are associated with an amount of user interaction with digital content that is below an interaction threshold;
generating, by the processing device, bids for the low-impression keywords using a model trained with an algorithm configured to train models based on data describing user interactions with the digital content and corresponding bids won for keywords and based on simulated winning training bids; and
submitting, by the processing device, the bids for the low-impression keywords to a search engine platform.