US 12,243,089 B2
Recommendation engine accounting for preference drift
Maharaj Mukherjee, Poughkeepsie, NY (US)
Assigned to Bank of America Corporation, Charlotte, NC (US)
Filed by Bank of America Corporation, Charlotte, NC (US)
Filed on Mar. 3, 2023, as Appl. No. 18/117,103.
Application 18/117,103 is a continuation of application No. 17/359,966, filed on Jun. 28, 2021, granted, now 11,636,530.
Prior Publication US 2023/0206303 A1, Jun. 29, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06Q 30/06 (2023.01); G06F 16/906 (2019.01); G06F 16/9536 (2019.01); G06Q 30/0601 (2023.01)
CPC G06Q 30/0631 (2013.01) [G06F 16/906 (2019.01); G06F 16/9536 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A computing platform comprising:
at least one processor;
a communication interface communicatively coupled to the at least one processor; and
memory storing computer-readable instructions that, when executed by the at least one processor, cause the computing platform to:
identify, at a first time and using a collaborative recommendation engine, a preference group for a user device, wherein identifying the preference group is based on a similarity of preferences between the preference group and a user of the user device, wherein the preference group corresponds to a first set of preferences at the first time, and wherein the collaborative recommendation engine is configured to output recommendation information based on:
historical user preference information of various user groups, and
data drift of the historical user preference information for each of the various user groups;
receive, at a second time later than the first time, an updated account access request from the user device;
identify, using the collaborative recommendation engine, the preference group and data drift corresponding to the preference group between the first time and the second time, wherein the data drift indicates that the preference group corresponds to a second set of preferences at the second time, and wherein identifying the data drift comprises:
identifying a different preference group that previously corresponded to the first set of preferences;
identifying that the different preference group transitioned from the first set of preferences to the second set of preferences; and
automatically predicting, based on the transition of the different preference group from the first set of preferences to the second set of preferences, the data drift indicating that the preference group corresponds to the second set of preferences at the second time;
automatically generate, based on the second set of preferences, updated recommendation information for the user device;
send, to the user device, the updated recommendation information and one or more commands directing the user device to display the updated recommendation information, wherein sending the one or more commands to the user device causes the user device to display the updated recommendation information;
receive user input indicating whether or not a user of the user device selected content based on the updated recommendation information; and
update the collaborative recommendation engine based on the user input, wherein the update is further based on:
identifying a number of users that have reacted to the updated recommendation information in a same way as the user,
comparing the number of users to a predetermined threshold,
based on identifying that the predetermined threshold is satisfied by the identified number of users, updating the collaborative recommendation engine based on the reaction, and
based on identifying that the predetermined threshold is not satisfied by the identified number of users, label a reaction of the user as an outlier in the collaborative recommendation engine.