Menu personalization
Provided are systems, methods, and computer-program products for generating a personalized item list. In various examples, a server computer on a network can receive a request that includes a user identifier. The computer can use the user identifier to look up a data model associated with the user identifier. The computer can further determine a geolocation, and use the geolocation to determine a list of items associated with an eatery at or near the geolocation. The computer can input the item list into the data model, for the data model to output a probability for each item, the probability indicating a likelihood that the user will select the item. The probabilities can be used to generate a personalized item list, which can be output onto the network for receipt by a computing device.
1. A method implemented by a server computer on a network, comprising:
receiving, over the network, input corresponding to a request for a personalized item list, wherein the input includes a user identifier associated with a user, and wherein the input is associated with a computing device on the network;
reading a data store using the user identifier, the data store storing a plurality of data models associated with a plurality of user identifiers, each data model being customized based on data associated with a respective user identifier, wherein reading the data store results in obtaining a particular data model associated with the user identifier;
determining a geolocation associated with the user identifier;
determining, using the geolocation, a list of menu items associated with an eatery, wherein the eatery is associated with the geolocation;
inputting the list of menu items associated with the eatery including a listing of edible ingredients of each menu item into the data model, wherein the data model generates and outputs a selection probability value for each menu item in the list of menu items, the selection probability indicating a likelihood of the user selecting a corresponding menu item based on the user's preference of the edible ingredients of the menu item;
sorting the list of menu items according to the selection probability value associated with each menu item;
generating the personalized item list based on the sorted list of menu items, the personalized item list including one or more of the menu items and corresponding edible ingredients;
associating numerical scores with each of the one or more menu items in the personalized item list;
determining a current time;
modifying the numerical scores according to the current time;
re-sorting the personalized item list according to the modified numerical scores; and
outputting the re-sorted personalized item list including the one or more menu items and corresponding edible ingredients onto the network for receipt by the computing device.
2. The method of claim 1 , wherein the user's preference is based on a time of day or a day of week.
3. The method of claim 1 , further comprising:
determining a recent item selection associated with the user identifier, wherein the recent item selection is determined from within a pre-determined time period preceding receipt of the input; and
modifying the numerical scores according to the recent item selection.
4. The method of claim 1 , further comprising:
reducing numerical scores for successive similar items in the personalized item list.
5. The method of claim 1 , wherein determining the geolocation and generating the personalized item list occurs at a point in time prior to receiving the input, such that receiving the input results in outputting of the personalized item list.
6. The method of claim 1 , wherein menu items that do not meet the user's dietary restriction are excluded from the personalized item list.
7. The method of claim 1 , wherein the list of menu items is associated with a particular eatery within a pre-determined distance from the geolocation.
8. The method of claim 1 , wherein each of the plurality of data models is trained using a machine learning method.
9. The method of claim 1 , further comprising:
receiving input corresponding to an item selection; and
adding the item selection to the data model.
10. The method of claim 1 , further comprising:
receiving input corresponding to a value associated with an item from the personalized item list, wherein the value is a positive value or a negative value; and
adding the value to the data model.
11. A server computer on a network, comprising: one or more processors; and
a non-transitory computer-readable medium including instructions that, when executed by the one or more processors, cause the one or more processors to perform operations including:
receiving, over the network, input corresponding to a request for a personalized item list, wherein the input includes a user identifier associated with a user, and wherein the input is associated with a computing device on the network;
reading a data store using the user identifier, the data store storing a plurality of data models associated with a plurality of user identifiers, each data model being customized based on data associated with a respective user identifier, wherein reading the data store results in obtaining a particular data model associated with the user identifier;
determining a geolocation associated with the user identifier;
determining, using the geolocation, a list of menu items associated with an eatery, wherein the eatery is associated with the geolocation;
inputting the list of menu items associated with the eatery including a listing of edible ingredients of each menu item into the data model, wherein the data model generates and outputs a selection probability value for each menu item in the list of menu items, the selection probability value indicating a likelihood of the user selecting a corresponding menu item based on the user's preference of the edible ingredients of the menu item;
sorting the list of menu items according to the selection probability value associated with each menu item;
generating the personalized item list based on the sorted list of menu items, the personalized item list including one or more of the menu items and corresponding edible ingredients;
associating numerical scores with each of the one or more menu items in the personalized item list;
determining a current time;
modifying the numerical scores according to the current time;
re-sorting the personalized item list according to the modified numerical scores; and
outputting the re-sorted personalized item list including the one or more menu items and corresponding edible ingredients onto the network for receipt by the computing device.
12. The server computer of claim 11 , wherein the user's preference is based on a time of day or a day of week.
13. The server computer of claim 11 , wherein the operations further include:
associating numerical scores with each item in the personalized item list; and
re-sorting the personalized item list according to the numerical scores.
14. The server computer of claim 11 , wherein determining the geolocation and generating the personalized item list occurs at a point in time prior to receiving the input, such that receiving the input results in outputting of the personalized item list.
15. A computer-program product tangibly embodied in a non-transitory machine-readable storage medium, including instructions that, when executed by one or more processors of a computing device on a network, cause the one or more processors to:
receive, over the network, input corresponding to a request for a personalized item list, wherein the input includes a user identifier associated with a user, and wherein the input is associated with a computing device on the network;
read a data store using the user identifier, the data store storing a plurality of data models associated with a plurality of user identifiers, each data model being customized based on data associated with a respective user identifier, wherein reading the data store results in obtaining a particular data model associated with the user identifier;
determine a geolocation associated with the user identifier;
determine, using the geolocation, a list of menu items associated with an eatery, wherein the eatery is associated with the geolocation;
input the list of menu items associated with the eatery including a listing of edible ingredients of each menu item into the data model, wherein the data model generates and outputs a selection probability value for each menu item in the list of menu items, the selection probability value indicating a likelihood of the user selecting a corresponding menu item based on the user's preference of the edible ingredients of the menu item;
sort the list of menu items according to the selection probability value associated with each menu item;
generate the personalized item list based on the sorted list of menu items, the personalized item list including one or more of the menu items and corresponding edible ingredients;
associate numerical scores with each of the one or more menu items in the personalized item list;
determine a current time;
modify the numerical scores according to the current time;
re-sort the personalized item list according to the modified numerical scores; and
output the re-sorted personalized item list including the one or more menu items and corresponding edible ingredients onto the network for receipt by the computing device.
16. The computer-program product of claim 15 , wherein the user's preference is based on a time of day or a day of week.
17. The computer-program product of claim 15 , wherein menu items that do not meet the user's dietary restriction are excluded from the personalized item list.