US 12,189,773 B2
Methods and apparatus for detecting whether a string of characters represents malicious activity using machine learning
Joshua Daniel Saxe, Wichita, KS (US)
Assigned to Invincea, Inc., Burlington, MA (US)
Filed by Invincea, Inc., Burlington, MA (US)
Filed on Nov. 10, 2023, as Appl. No. 18/506,962.
Application 18/506,962 is a continuation of application No. 18/068,090, filed on Dec. 19, 2022, granted, now 11,853,427.
Application 18/068,090 is a continuation of application No. 17/125,280, filed on Dec. 17, 2020, granted, now 11,544,380, issued on Jan. 3, 2023.
Application 17/125,280 is a continuation of application No. 16/425,115, filed on May 29, 2019, granted, now 10,878,093, issued on Dec. 29, 2020.
Application 16/425,115 is a continuation of application No. 15/630,495, filed on Jun. 22, 2017, granted, now 10,318,735, issued on Jun. 11, 2019.
Claims priority of provisional application 62/353,286, filed on Jun. 22, 2016.
Prior Publication US 2024/0152617 A1, May 9, 2024
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 21/56 (2013.01); G06N 3/04 (2023.01); G06N 3/045 (2023.01); G06N 3/08 (2023.01); G06N 3/084 (2023.01); G06N 5/01 (2023.01); G06N 7/01 (2023.01); H04L 9/40 (2022.01)
CPC G06F 21/567 (2013.01) [G06F 21/562 (2013.01); G06N 3/04 (2013.01); G06N 3/045 (2023.01); G06N 3/08 (2013.01); G06N 3/084 (2013.01); H04L 63/1416 (2013.01); G06F 2221/032 (2013.01); G06N 5/01 (2023.01); G06N 7/01 (2023.01)] 18 Claims
OG exemplary drawing
 
1. An apparatus, comprising:
a memory; and
one or more processors operatively coupled to the memory, the one or more processors configured to:
receive a string associated with an artifact;
convert each character in the string into a character vector to generate a set of character vectors;
apply a convolution matrix to the set of character vectors to define at least a portion of a feature vector;
provide the feature vector as an input to a machine learning model configured to receive the feature vector and generate as an output a classification indicating a maliciousness associated with the artifact based on the feature vector; and
detect the maliciousness associated with the artifact based on the output of the machine learning model.