US 11,989,326 B2
Programmable feature extractor with anonymization
Joseph H. Levy, Farmington, UT (US); Kenneth D. Ray, Seattle, WA (US); and Joshua Daniel Saxe, Wichita, KS (US)
Assigned to Sophos Limited, Abingdon (GB)
Filed by Sophos Limited, Abingdon (GB)
Filed on Mar. 30, 2021, as Appl. No. 17/217,894.
Prior Publication US 2022/0318429 A1, Oct. 6, 2022
Int. Cl. G06F 21/62 (2013.01); G06F 21/53 (2013.01); G06N 20/00 (2019.01); H04L 67/06 (2022.01)
CPC G06F 21/6254 (2013.01) [G06F 21/53 (2013.01); G06N 20/00 (2019.01); H04L 67/06 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
extracting, by a compute instance, a feature of a data instance accessed by the compute instance;
generating, by the compute instance, an anonymized feature value for the feature of the data instance;
including, by the compute instance, the anonymized feature value in a feature vector corresponding to the data instance;
transmitting, by the compute instance, the feature vector to a server-based computing system; and
receiving, by the compute instance, data representative of consent by a user of the compute instance for the compute instance to extract the feature, generate the anonymized feature value, include the anonymized feature value in the feature vector, and transmit the feature vector to the server-based computing system;
wherein the extracting of the feature, the generating of the anonymized feature value, the including of the anonymized feature value in the feature vector, and the transmitting of the feature vector to the server-based computing system are performed in response to the receiving the data representative of the consent.