US 12,242,604 B2
Method for preventing data leakage to machine learning engines available in electronic device
Rahul Agrawal, Gwalior (IN); Vipul Gupta, Hisar (IN); Saurabh Kumar, Patna Sahib (IN); Ankur Agrawal, Agra (IN); and Nitesh Goyal, Ghaziabad (IN)
Assigned to SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR)
Filed by SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR)
Filed on Dec. 30, 2021, as Appl. No. 17/566,369.
Application 17/566,369 is a continuation of application No. PCT/KR2021/018542, filed on Dec. 8, 2021.
Claims priority of application No. 202041055130 (IN), filed on Dec. 18, 2020.
Prior Publication US 2022/0198006 A1, Jun. 23, 2022
Int. Cl. G06F 21/55 (2013.01); G06N 20/00 (2019.01)
CPC G06F 21/556 (2013.01) [G06N 20/00 (2019.01); G06F 2221/034 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method for preventing data leakage in an electronic device, the method comprising:
detecting a data request from a first machine learning (ML) engine of a plurality of ML engines that requests at least one framework application of a plurality of framework applications to provide data;
identifying the data that is generated by the at least one framework application in response to the data request from the first ML engine;
creating a plurality of data blocks based on the data generated by the at least one framework application, a first category of the first ML engine that is determined based on a communication pattern between the first ML engine and the at least one framework application, and a first tag associated with the first ML engine and the at least one framework application;
determining whether the plurality of data blocks are valid to share with the first ML engine using an activity block chain associated with each of the plurality of framework applications;
based on the plurality of data blocks being valid, sharing the plurality of data blocks with the first ML engine, as a valid set of data blocks; and
based on the plurality of data blocks not being valid, discarding the plurality of data blocks, as an invalid set of data blocks, not to be shared with the first ML engine.