US 11,748,770 B1
System, method, and computer program for using shared customer data and artificial intelligence to predict customer classifications
Sarit Chehanowitz, Tel Aviv (IL); Liat Taub Bahar, Kfar Sabba (IL); and Shmuel Ur, Shorashim (IL)
Assigned to AMDOCS DEVELOPMENT LIMITED, Limassol (CY)
Filed by Amdocs Development Limited, Limassol (CY)
Filed on Mar. 30, 2020, as Appl. No. 16/835,010.
Int. Cl. G06Q 30/0201 (2023.01); G06F 16/906 (2019.01); G06F 16/9536 (2019.01)
CPC G06Q 30/0201 (2013.01) [G06F 16/906 (2019.01); G06F 16/9536 (2019.01)] 6 Claims
OG exemplary drawing
 
1. A non-transitory computer readable medium storing computer code executable by a processor to perform a method comprising:
receiving, by a first system of a first business entity from a third-party system to the first system of the first business entity and a second system of a second business entity, an artificial intelligence model configured to process a first input to create a predictor capable of predicting classifications from customer data, wherein the artificial intelligence model is generated using machine learning performed on output of a secure multi-party computation applied to:
a first database schema according to which first customer data is stored in a first database by the first system of the first business entity, and
a second database schema according to which second customer data is stored in a second database by the second system of the second business entity,
wherein the secure multi-party computation is applied by the third-party system to the first system of the first business entity and the second system of the second business entity,
wherein the secure multi-party computation jointly cryptographically computes a function over the first database schema and the second database schema,
wherein the output of the secure multi-party computation is a subset database schema that is common to both the first database schema and the second database schema, and
wherein the artificial intelligence model is generated by the third party system from the output of the secure multi-party computation such that the artificial intelligence model is generated without sharing the first customer data with the second business entity and without sharing the second customer data with the first business entity;
executing, by the first system of the first business entity, the artificial intelligence model on the first schema, the first customer data, and customer classifications to generate a first predictor, the first predictor configured to receive a second input and process the second input to predict a classification for the second input,
wherein the third-party system receives the first predictor from the first system for generating a translation for the first predictor between the first database schema according to which the first customer data is stored and the second database schema according to which the second customer data is stored;
distributing, by the first system of the first business entity through the third party system, the first predictor with the translation to the second system of the second business entity for use by the second system of the second business entity to predict at least one classification for the second customer data.