US 11,704,717 B2
Item affinity processing
Itamar David Laserson, Givat Shmuel (IL); Rotem Chudin, Kfar-Saba (IL); Julie Dvora Katz Ohayon, Ashdod (IL); and Moshe Shaharur, Tel-Aviv (IL)
Assigned to NCR Corporation, Atlanta, GA (US)
Filed by NCR Corporation, Atlanta, GA (US)
Filed on Sep. 24, 2020, as Appl. No. 17/30,554.
Prior Publication US 2022/0092670 A1, Mar. 24, 2022
Int. Cl. G06Q 30/06 (2023.01); G06Q 30/0601 (2023.01); G06N 20/00 (2019.01); G06F 40/242 (2020.01); G06Q 10/0875 (2023.01)
CPC G06Q 30/0631 (2013.01) [G06F 40/242 (2020.01); G06N 20/00 (2019.01); G06Q 10/0875 (2013.01); G06Q 30/0603 (2013.01)] 13 Claims
OG exemplary drawing
 
1. A method, comprising:
providing executable instructions to a processor of a server and causing the processor to execute the executable instructions and perform operations comprising:
mapping item codes of a product catalogue to item vectors plotted in multidimensional space based on contexts of the item codes within a transaction history, wherein mapping further includes deriving the contexts based on patterns detected across transactions of the transaction history for each item code;
assigning each item code to a corresponding item vector;
providing substitute item codes for a given item code that is out-of-stock based on similarity values calculated from distances within the multidimensional space between first vectors associated with the substitute item codes and a second vector associated with the given item code, wherein providing the substitute item codes further includes adjusting the substitute item codes based on feedback associated with previous substitute item codes for the given item code that was purchased by a customer or that was not purchased by the customer;
providing the substitute item codes in real time to a transaction interface, a picking system, or a management application for a current transaction being processed, restocking being processed, or a current order being processed; and
providing and processing the method as a data driven item substitution cloud service predicter to the transaction interface during transactions, the picking system during order fulfillment, and the management application during restocking activities at a store, wherein the data driven item substitution cloud service predicter is provided as a software-as-a-service to the transaction interface, the picking system, and the management application through an application programming interface.