Cognitive information security using a behavioral recognition system

• US 9,507,768 B2
• Filed: 08/11/2014
• Issued: 11/29/2016
• Est. Priority Date: 08/09/2013
• Status: Active Grant

• Current Assignee: Intellective Ai, Inc.
• Original Assignee: Behavioral Recognition Systems Incorporated
• Inventors: Cole, Curtis Edward Jr., Falcon, Cody Shay, Konosky, Benjamin A., Morgan, Charles Richard, Poffenberger, Aaron, Nguyen, Thong Toan, Cobb, Wesley Kenneth, Seow, Ming-Jung
• Primary Examiner(s): Hirl, Joseph P
• Assistant Examiner(s): GUNDRY, STEPHEN T
• Application Number:
  US14/457,060
• Publication Number:
  US 20150047040A1
• Time in Patent Office: 841 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 40/226 Validation
• CPC Class Codes: G06F 40/242 Dictionaries
• CPC Class Codes: G06F 40/247 Thesauruses; Synonyms
• CPC Class Codes: G06F 40/253 Grammatical analysis; Style critique
• CPC Class Codes: G06F 40/284 Lexical analysis, e.g. tokenisation or collocates
• CPC Class Codes: G06F 40/289 Phrasal analysis, e.g. finite state techniques or chunking
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06F 40/40 Processing or translation of natural language natural language analysis G06F40/20; semantic analysis G06F40/30
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 3/0409 Adaptive resonance theory [ART] networks
• CPC Class Codes: G06N 3/042 Knowledge-based neural networks; Logical representations of neural networks
• CPC Class Codes: G06N 3/088 Non-supervised learning, e.g. competitive learning
• CPC Class Codes: G06N 5/022 Knowledge engineering; Knowledge acquisition
• CPC Class Codes: H04L 63/1408 by monitoring network traffic monitoring network traffic per se H04L43/00
• CPC Class Codes: H04L 63/1425 Traffic logging, e.g. anomaly detection

Abstract

Embodiments presented herein describe a method for processing streams of data of one or more networked computer systems. According to one embodiment of the present disclosure, an ordered stream of normalized vectors corresponding to information security data obtained from one or more sensors monitoring a computer network is received. A neuro-linguistic model of the information security data is generated by clustering the ordered stream of vectors and assigning a letter to each cluster, outputting an ordered sequence of letters based on a mapping of the ordered stream of normalized vectors to the clusters, building a dictionary of words from of the ordered output of letters, outputting an ordered stream of words based on the ordered output of letters, and generating a plurality of phrases based on the ordered output of words.

Claims

1. A computer-implemented method for processing streams of information security data from one or more networked computer systems, the method comprising:

receiving, by a machine learning engine executing on one or more computing systems, an ordered stream of normalized vectors corresponding to information security data obtained from one or more sensors monitoring a computer network;

generating a neuro-linguistic model of the information security data by;

clustering the ordered stream of vectors and assigning a letter to each cluster,

outputting an ordered sequence of letters based on a mapping of the ordered stream of normalized vectors to the clusters,

building a dictionary of words from of the ordered output of letters,

outputting an ordered stream of words based on the ordered output of letters, and

generating a plurality of phrases based on the ordered output of words;

evaluating a current observation of letters, words, or phrases generated from one or more subsequent normalized vectors received from the ordered stream input to the neuro-linguistic model to determine a measure of unusualness for the current observation of letters, words, or phrases;

generating an alert when the measure of unusualness determined for the current observation of letters, words, or phrases exceeds a specified threshold, wherein the alert corresponds to activity occurring within the computer network which resulted in the current observation of letters, words, or phrases;

transmitting the alert to a management console; and

dynamically updating the neuro-linguistic model based on the one or more subsequent normalized vectors from the ordered stream input to the neuro-linguistic model.

2. The method of claim 1, wherein generating the alert when the measure of unusualness for the current observation of letters, words, phrases, differs from the neuro-linguistic model of the information security data comprises determining, for at least a first, letter, word, or phrase in the current observation, the unusualness score based on a frequency of letters, words, and phrases, respectively, in the letters, words, and phrases of the neuro-linguistic model;

and

generating the alert when the unusualness score for at least one of the letters, words, or phases in the current observation exceeds the specified threshold.

3. The method of claim 1, wherein each value of the information security data associated with the normalized vectors is normalized to a value within a range of 0 to 1, inclusive.

4. The method of claim 1, wherein the information security data comprises network packet traffic information, disk mount event information, or security log data.

5. The method of claim 1, wherein building the dictionary comprises building a dictionary of words up to a maximum length of letters.

6. A non-transitory computer-readable storage medium storing instructions, which, when executed on a processor, performs an operation for processing streams of data of one or more networked computer systems, the operation comprising:

receiving an ordered stream of normalized vectors corresponding to information security data obtained from one or more sensors monitoring a computer network;

generating a neuro-linguistic model of the information security data by;

clustering the ordered stream of vectors and assigning a letter to each cluster,

outputting an ordered sequence of letters based on a mapping of the ordered stream of normalized vectors to the clusters,

building a dictionary of words from of the ordered output of letters,

outputting an ordered stream of words based on the ordered output of letters, and

generating a plurality of phrases based on the ordered output of words;

evaluating a current observation of letters, words, or phrases generated from one or more subsequent normalized vectors received from the ordered stream input to the neuro-linguistic model to determine a measure of unusualness for the current observation of letters, words, or phrases;

generating an alert when the measure of unusualness determined for the current observation of letters, words, or phrases exceeds a specified threshold, wherein the alert corresponds to activity occurring within the computer network which resulted in the current observation of letters, words, or phrases;

transmitting the alert to a management console; and

dynamically updating the neuro-linguistic model based on the one or more subsequent normalized vectors from the ordered stream input to the neuro-linguistic model.

7. The non-transitory computer-readable storage medium of claim 6, wherein generating the alert when the measure of unusualness for the current observation of letters, words, phrases, differs from the neuro-linguistic model of the information security data comprises determining, for at least a first, letter, word, or phrase in the current observation, the unusualness score based on a frequency of letters, words, and phrases, respectively, in the letters, words, and phrases of the neuro-linguistic model;

and

generating the alert when the unusualness score for at least one of the letters, words, or phases in the current observation exceeds the specified threshold.

8. The non-transitory computer-readable storage medium of claim 6, wherein the each value of the information security data associated with the normalized vectors is normalized to a value within a range of 0 to 1, inclusive.

9. The non-transitory computer-readable storage medium of claim 6, wherein the information security data comprises network packet traffic information, disk mount event information, or security log data.

10. The non-transitory computer-readable storage medium of claim 6, wherein building the dictionary comprises building a dictionary of words up to a maximum length of letters.

11. A system, comprising:

a processor; and

a memory storing one or more application programs configured to perform an operation for processing streams of data of one or more networked computer systems, the operation comprising;

receiving an ordered stream of normalized vectors corresponding to information security data obtained from one or more sensors monitoring a computer network,

generating a neuro-linguistic model of the information security data by;

clustering the ordered stream of vectors and assigning a letter to each cluster;

outputting an ordered sequence of letters based on a mapping of the ordered stream of normalized vectors to the clusters;

building a dictionary of words from of the ordered output of letters;

outputting an ordered stream of words based on the ordered output of letters; and

generating a plurality of phrases based on the ordered output of words,

evaluating a current observation of letters, words, or phrases generated from one or more subsequent normalized vectors received from the ordered stream input to the neuro-linguistic model to determine a measure of unusualness for the current observation of letters, words, or phrases,

generating an alert when the measure of unusualness determined for the current observation of letters, words, or phrases exceeds a specified threshold, wherein the alert corresponds to activity occurring within the computer network which resulted in the current observation of letters, words, or phrases,

transmitting the alert to a management console, and

dynamically updating the neuro-linguistic model based on the one or more subsequent normalized vectors from the ordered stream input to the neuro-linguistic model.

12. The system of claim 11, wherein generating the alert when the measure of unusualness for the current observation of letters, words, phrases, differs from the neuro-linguistic model of the information security data comprises determining, for at least a first, letter, word, or phrase in the current observation, the unusualness score based on a frequency of letters, words, and phrases, respectively, in the letters, words, and phrases of the neuro-linguistic model;

and

generating the alert when the unusualness score for at least one of the letters, words, or phases in the current observation exceeds the specified threshold.

13. The system of claim 11, wherein the each value of the information security data associated the normalized vectors is normalized to a value within a range of 0 to 1, inclusive.

14. The system of claim 11, wherein the information security data comprises network packet traffic information, disk mount event information, or security log data.

15. The system of claim 11, wherein building the dictionary comprises building a dictionary of words up to a maximum length of letters.