Back to results
US 11,238,495 B2Granted
Method and system for data driven personalization
AssigneeWalmart Apollo, LLC
InventorsKannan Achan, Venkata Syam Prakash Rapaka, Jeremy King
Life of this patent
Estimating the term…
Abstract
A system and method for providing a customized user experience is presented. A system can include one or more processing modules and one more non-transitory storage modules. The usage history of a user can be tracked at an eCommerce retailer. Static information about the user can also be tracked. Thereafter, using the usage history, the static information, and the available promotions of the retailer, a probability of conversion can be calculated for each promotion. Thereafter, the promotion being displayed to the user can be based on the probability of conversion. Other embodiments are also disclosed herein.
Claims22
A system comprising:
- one or more user input devices;
- one or more processors;
- and
- one or more non-transitory storage devices storing computing instructions configured to run on the one or more processors and perform:
- tracking, using the one or more processors, usage history of a user comprising:
- interactions of the user with an eCommerce retailer, the interactions collected across multiple electronic devices of the user;
- and
- views and purchases of the user collected at a website of the eCommerce retailer stored in one or more electronic cookies;
- storing the usage history of the user at the one or more non-transitory storage devices;
- tracking, using the one or more processors, static information about the user;
- storing the static information about the user at the one or more non-transitory storage devices;
- aggregating the usage history of the user and the static information about the user with other usage history of a plurality of other users and other static information about the plurality of other users;
- creating training data from the usage history of the user, the static information about the user, the other usage history of the plurality of other users and the other static information about the plurality of other users, as aggregated;
- accessing the one or more non-transitory storage devices to retrieve the training data;
- training a machine learning algorithm on the training data;
- for each respective promotion in a plurality of promotions:
- using the machine learning algorithm, as trained, to determine a respective probability of conversion for the user to be converted by the respective promotion of the plurality of promotions;
- and
- assigning a respective weight to the respective promotion of the plurality of promotions based on the respective probability of conversion for the user to be converted by the respective promotion of the plurality of promotions;
- causing to be displayed, on an electronic device of the user and using one or more markup languages, at least one respective promotion from the plurality of promotions based on its respective weight, as assigned, for the at least one respective promotion, wherein:
- the at least one respective promotion from the plurality of promotions are randomly selected from the plurality of promotions;
- and
- a number of times a respective one of the at least one respective promotion from the plurality of promotions is caused to be displayed on the electronic device of the user is directly proportional to its weight, as assigned, as compared with weights, as assigned, of other respective promotions of the at least one respective promotion;
- after a predetermined time period has elapsed, updating the usage history of the user;
- re-training the machine learning algorithm, as trained, on the usage history of the user, as updated;
- for each respective promotion in the plurality of promotions:
- using the machine learning algorithm, as trained and re-trained, to determine a new respective probability of conversion for the user to be converted by the respective promotion of the plurality of promotions;
- and
- assigning a new respective weight to the respective promotion of the plurality of promotions based on the new respective probability of conversion for the user to be converted by the respective promotion of the plurality of promotions;
- and
- causing to be displayed, on the electronic device of the user, the at least one respective promotion from the plurality of promotions based on the new respective weight, as assigned, of the at least one respective promotion.
28 more elements
A method comprising:
- tracking, using a processor, usage history of a user comprising:
- interactions of the user with an eCommerce retailer, the interactions collected across multiple electronic devices of the user;
- and
- views and purchases of the user collected at a website of the eCommerce retailer stored in one or more electronic cookies;
- storing the usage history of the user at one or more non-transitory storage devices;
- tracking, using the processor, static information about the user;
- storing the static information about the user at the one or more non-transitory storage devices;
- aggregating the usage history of the user and the static information about the user with other usage history of a plurality of other users and other static information about the plurality of other users;
- creating training data from the usage history of the user, the static information about the user, the other usage history of the plurality of other users and the other static information about the plurality of other users, as aggregated;
- accessing the one or more non-transitory storage devices to retrieve the training data;
- training a machine learning algorithm on the training data;
- for each respective promotion in a plurality of promotions:
- using the machine learning algorithm, as trained, to determine a respective probability of conversion for the user to be converted by the respective promotion of the plurality of promotions;
- and
- assigning a respective weight to the respective promotion of the plurality of promotions based on the respective probability of conversion for the user to be converted by the respective promotion of the plurality of promotions;
- causing to be displayed, on an electronic device of the user and using one or more markup languages, at least one respective promotion from the plurality of promotions based on its respective weight, as assigned, for the at least one respective promotion, wherein:
- the at least one respective promotion from the plurality of promotions are randomly selected from the plurality of promotions;
- and
- a number of times a respective one of the at least one respective promotion from the plurality of promotions is caused to be displayed on the electronic device of the user is directly proportional to its weight, as assigned, as compared with weights, as assigned, of other respective promotions of the at least one respective promotion;
- after a predetermined time period has elapsed, updating the usage history of the user;
- re-training the machine learning algorithm, as trained, on the usage history of the user, as updated;
- for each respective promotion in the plurality of promotions:
- using the machine learning algorithm, as trained and re-trained, to determine a new respective probability of conversion for the user to be converted by the respective promotion of the plurality of promotions;
- and
- assigning a new respective weight to the respective promotion of the plurality of promotions based on the new respective probability of conversion for the user to be converted by the respective promotion of the plurality of promotions;
- and
- causing to be displayed, on the electronic device of the user, the at least one respective promotion from the plurality of promotions based on the new respective weight, as assigned, of the at least one respective promotion.
24 more elements
Dependencies are inferred from the claim text.