US 11,669,784 B2
Content provider recommendations to improve targetting and other settings
Mahesh Keralapura Manjunatha, Mountain View, CA (US); and Chiu Wah So, Mountain View, CA (US)
Assigned to GOOGLE LLC, Mountain View, CA (US)
Appl. No. 16/756,347
Filed by GOOGLE LLC, Mountain View, CA (US)
PCT Filed Dec. 11, 2019, PCT No. PCT/US2019/065733
§ 371(c)(1), (2) Date Apr. 15, 2020,
PCT Pub. No. WO2021/118559, PCT Pub. Date Jun. 17, 2021.
Prior Publication US 2021/0184976 A1, Jun. 17, 2021
Int. Cl. H04L 47/215 (2022.01); H04L 47/11 (2022.01); H04L 65/612 (2022.01); H04L 67/53 (2022.01); G06F 18/22 (2023.01)
CPC H04L 47/215 (2013.01) [G06F 18/22 (2023.01); H04L 47/11 (2013.01); H04L 65/612 (2022.05); H04L 67/53 (2022.05)] 20 Claims
OG exemplary drawing
 
1. A method for pruning selection criteria from a set of selection criteria based on document space associations, comprising:
constructing, by a data processing system including one or more processors from a content provider computing system, a first data structure indicating associated predicted requests for a first retrieval token of a plurality of retrieval tokens;
comparing, by the data processing system, the first data structure with a second data structure indicating associated predicted requests for a second retrieval token of the plurality of retrieval tokens to calculate a similarity value;
determining, by the data processing system, the similarity value is greater than a predetermined threshold;
removing, by the data processing system, the first retrieval token and the second retrieval token from the plurality of retrieval tokens to create a pruned plurality of retrieval tokens responsive to determining the similarity value is greater than a predetermined threshold; and
providing, by the data processing system, the pruned plurality of retrieval tokens to the content provider computing device.