US 12,014,394 B2
Apparatus, computer-implemented method, and computer program product for programmatically selecting a user survey data object from a set of user survey data objects and for selecting ranking model(s) for utilization based on survey engagement data associated with the selected user survey data object
Pradyumna Siddhartha, Marina Del Rey, CA (US); Mohammad Sabah, San Pedro, CA (US); Qiqi Ai, Gardena, CA (US); Mo Lin, Los Angeles, CA (US); and Aakash Pathak, Marina Del Rey, CA (US)
Assigned to THRIVE MARKET, INC., Los Angeles, CA (US)
Filed by Thrive Market, Inc., Los Angeles, CA (US)
Filed on Mar. 7, 2023, as Appl. No. 18/179,649.
Application 18/179,649 is a continuation of application No. 17/139,904, filed on Dec. 31, 2020, granted, now 11,631,106.
Prior Publication US 2023/0368241 A1, Nov. 16, 2023
Int. Cl. G06Q 30/0242 (2023.01); G06Q 30/0217 (2023.01); G06Q 30/0251 (2023.01)
CPC G06Q 30/0245 (2013.01) [G06Q 30/0218 (2013.01); G06Q 30/0269 (2013.01)] 20 Claims
OG exemplary drawing
 
1. An apparatus comprising at least one processor and at least one memory storing instructions that, with at least one processor, cause the apparatus to:
determine at least one model ranking metric value for each candidate trained recommendation ranking model of a set of candidate trained recommendation ranking models,
wherein the set of candidate trained recommendation ranking models comprises a plurality of distinct model types,
wherein model ranking metric value of the at least one model ranking metric value corresponds to a particular inquiry response data object of a set of inquiry response data objects;
determine, based at least in part on the model ranking metric value for each candidate trained recommendation ranking model, a highest model ranking metric value, the highest model ranking metric value corresponding to a particular trained recommendation ranking model from the set of candidate trained recommendation ranking models; and
link the particular trained recommendation ranking model corresponding to the highest model ranking metric value with at least the particular inquiry response data object.