US 12,299,701 B2
Utilizing lifetime values of users to select content for presentation in a messaging system
Jean Luo, Seattle, WA (US); and Zhehao Zhou, Santa Monica, CA (US)
Assigned to Snap Inc., Santa Monica, CA (US)
Filed by Snap Inc., Santa Monica, CA (US)
Filed on Sep. 29, 2021, as Appl. No. 17/489,102.
Claims priority of provisional application 63/085,946, filed on Sep. 30, 2020.
Prior Publication US 2022/0101349 A1, Mar. 31, 2022
Int. Cl. G06Q 30/0201 (2023.01); G06Q 30/0204 (2023.01); G06Q 30/0241 (2023.01); G06T 19/00 (2011.01)
CPC G06Q 30/0201 (2013.01) [G06Q 30/0205 (2013.01); G06Q 30/0277 (2013.01); G06T 19/006 (2013.01)] 16 Claims
OG exemplary drawing
 
1. A method, comprising:
determining, by one or more hardware processors provided by a messaging server system, monthly active users (MAU) and penetration for global users and users in a specific country;
predicting, by the one or more hardware processors of the messaging server system, monetization values for the users in the specific country;
determining, by a first hardware processor of the messaging server system, a lifetime value of the users in the specific country based at least in part on the monetization values;
selecting, by a second hardware processor accessing a shared memory of the messaging server system, at least one augmented reality (AR) content generator based at least in part on the determined lifetime value of the users in the specific country, the users in the specific country comprising a set of users based on user interactions of augmented reality content generators received from multiple clients devices over a network, the network comprising the Internet, the selecting, using the second hardware processor accessing the shared memory, the at least one AR content generator being further based on determining that a percentage of monetized users on a daily basis is lower than a percentage of the users of the at least one AR content generator;
performing, by the one or more hardware processors of the messaging server system, a validation process of a model to determine the lifetime value of the users in the specific country, the validation process comprising:
determining a frequency of repeat transactions based on a number of users and a number of calibration period transactions;
determining a number of purchases in a holdout period and a number of predicted purchases based on an average of purchases in the holdout period and a number of purchases in a calibration period;
determining, based on the number of purchases in the holdout period and the number of predicted purchases, whether a cumulative error rate is below a predetermined percentage; and
indicating that the model is accurate based at least in part on determining that the cumulative error rate is below the predetermined percentage;
causing, at a client device over the Internet, display of the at least one AR content generator on a display of the client device; and
displaying, at the display of the client device, the at least one AR content generator.