Granted Patent B2

US 12,373,557 · App. 18/166,654 · Granted Jul 29, 2025

Inventor: Christopher G. Coulter (Stuart, FL)

Assignee: Auguria, Inc.

G06F21/554G06N3/091G06F2221/034

Abstract

Claims (52)

1. A method, comprising:

receiving, from a source from a plurality of sources, data associated with an event at the source, the data including an event timestamp and an event data-frame;

defining a standardized event based on the event by parsing the data to map predefined data fields within the event data-frame based on predefined standardization rules;

tokenizing the standardized event to produce a tokenized event;

providing the tokenized event to a word embedding model to define a vector representation of the event having a predetermined length, the vector representation of the event being a representation of the standardized event;

calculating a representative identifier for the event by providing the representation of the standardized event as an input to a predefined function;

defining an identifier for the event based on the representative identifier;

calculating a distance of the identifier to an identifier associated with each event cohort from a plurality of event cohorts;

assigning the event to an event cohort from the plurality of event cohorts based on the distance meeting a criterion of that cohort; and

identifying an anomalous event based on the distance exceeding a predetermined threshold.

2. The method of claim 1 , further comprising:

identifying a set of uncommon events based on a distance associated with a representative identifier of each event from the set of uncommon events not meeting a criterion of an event cohort from the plurality of event cohorts.

3. The method of claim 2 , wherein the identifying the set of uncommon events includes identifying the set of uncommon events based on metadata associated with each event from the set of uncommon events.

4. The method of claim 1 , wherein the calculating the distance is based on at least one of a cosine similarity, a hamming distance, a nearest neighbor search or a Euclidean distance.

5. The method of claim 1 , wherein the calculating the distance is based on a locality sensitive hashing (LSH) function.

6. The method of claim 1 , wherein the predefined function is one of a hash function, a vector function, a second tokenization function, or a feature function.

7. The method of claim 1 , wherein the defining the identifier is based on at least one of a recurrent neural network (RNN) or a transformer.

8. A method, comprising:

receiving, from a plurality of sources, data associated with a plurality of events at the plurality of sources;

standardizing the data based on a set of predefined standardization rules to define standardized data;

tokenizing the standardized data to produce tokenized data;

providing the tokenized data to a word embedding model to define a vector representation of the standardized data having a predetermined length;

transforming, by a machine learning model, the vector representation to define transformed data;

sorting the transformed data using a coarse sorting process to define coarse sorted data;

sorting the coarse sorted data using a fine sorting process to define fine sorted data, the fine sorted data defining a plurality of distances between data points of the fine sorted data; and

based on the fine sorted data, identifying an anomalous event from the plurality of events.

9. The method of claim 8 , wherein the coarse sorting process includes using a locality sensitive hashing (LSH) function.

10. The method of claim 8 , wherein the method further includes:

applying a locality sensitive hashing (LSH) function on the standardized data prior to transforming the vector representation.

11. The method of claim 8 wherein the coarse sorting process includes using a WordPiece tokenization algorithm.

12. The method of claim 8 , wherein the fine sorting process includes using at least one of a cosine similarity, a hamming distance, a Euclidean distance, or a nearest neighbor search.

13. The method of claim 8 , wherein the machine learning model includes at least one of a neural network, reinforcement learning, a Bayesian network, a random forest, a support vector machine or a decision tree.

14. The method of claim 8 , further comprising:

defining a plurality of scores based on the plurality of distances, each score from the plurality of scores associated with a respective event from the plurality of events.

15. The method of claim 14 , wherein a score from the plurality of scores is associated with a distance from the plurality of distances and between a first event from the plurality of events and a second event from the plurality of events and in the same event cohort as the first event.

16. The method of claim 14 , wherein a score from the plurality of scores is associated with a distance from the plurality of distances and between an event from the plurality of events and a cluster different than an event cohort associated with the event.

17. The method of claim 8 , further comprising:

performing a remedial action based on identifying the anomalous event, the remedial action being at least one of identifying the anomalous event as indicative of malware, quarantining a file associated with the anomalous event, or restricting access to a device associated with the anomalous event.

18. A non-transitory processor-readable medium storing code representing instructions to be executed by one or more processors, the instructions comprising code to cause the one or more processors to:

receive, from a plurality of sources, data associated with a plurality of events at the plurality of sources;

standardize the data based on a set of predefined standardization rules to define standardized data;

tokenize the standardized data to produce tokenized data;

provide the tokenized data to a word embedding model to define a vector representation of the standardized data having a predetermined length;

transform, by a machine learning model, the vector representation to define transformed data;

sort the transformed data using a coarse sorting process to define coarse sorted data;

sort the coarse sorted data using a fine sorting process to define fine sorted data, the fine sorted data defining a plurality of distances between data points of the fine sorted data; and

based on a distance from the plurality of distances and associated with an event from the plurality of events exceeding a threshold, quarantine an artifact associated with the event from the plurality of events.

19. The non-transitory processor-readable medium of claim 18 , wherein the code further comprises code to cause the one or more processors to:

identify a set of uncommon events based on a distance from the plurality of distances for each event from the set of uncommon events not meeting a criterion.

20. The non-transitory processor-readable medium of claim 19 , wherein the criterion is associated with an event cohort from a plurality of event cohorts associated with the plurality of distances.

21. The non-transitory processor-readable medium of claim 19 , wherein identifying the set of uncommon events includes identifying the set of uncommon events based on metadata associated with each event from the set of uncommon events.

22. The non-transitory processor-readable medium of claim 18 , wherein the coarse sorting process includes using a locality sensitive hashing (LSH) function.

Assignments (1)

Continuity (2)

References Cited (30)