Luke Higgins

US Patent:

20210067532, Mar 4, 2021

Filed:

Sep 4, 2019

Appl. No.:

16/560148

Inventors:

- New York NY, US
Luke A. Higgins - Silver Spring MD, US

International Classification:

H04L 29/06
G06K 9/62
G06Q 10/06

Abstract:

A method of detecting and responding to anomalous activity within a system involves, based upon pivot feature information, for a snapshot interval, generating a frequency structure interrelating a pivot feature, a binning feature and counts, using the frequency structure, generating a graphical image of a pre-specified width and height in pixels, and wherein the pre-specified width corresponds to a cyclical repeating interval made up of multiple bins, the graphical image having been generated such that a bin of the multiple bins, is the bin having a highest count, and the bin is scaled to the pre-specified height, and counts of all other bins in the interval are scaled relative to that highest count, graphically comparing the generated graphical image to an immediately preceding graphical image for similarity, and if a result of the comparison fails to satisfy a pre-specified similarity threshold, automatically triggering an appropriate anomaly detection-based follow-on action.