Systems and methods for an at-risk system identification via analysis of online hacker community discussions

Patent metadata

Field	Value
Publication number	US-12380221-B2
Application number	US-202318343138-A
Country	US
Kind code	B2
Filing date	Jun 28, 2023
Priority date	May 15, 2019
Publication date	Aug 5, 2025
Grant date	Aug 5, 2025

Abstract

Various embodiments of systems and methods for an at-risk system identification via analysis of discussions from various online hacker communities are disclosed herein.

First claim

What is claimed is:

1. A system for computer-implemented identification of at-risk systems, comprising:
a processor in communication with a tangible storage medium storing instructions that are executed by the processor such that the processor:
accesses input data including a hacker discussion discussing a vulnerability;
constructs arguments for a given query associated with a set of at-risk system components using elements in a predetermined knowledge base, the arguments for and against each of the set of components of a system being of risk to exploitation based on a hacker communication to constrain the set of components to a reduced set of components; and
executes a machine learning model trained to use as input the reduced set of components and the query to identify a system associated with the reduced set of components that is susceptible to the vulnerability and at risk to a cyber-attack, identification of the system being prior to the cyber-attack occurring.

2. The system of claim 1 , wherein the set of system components includes possible platform components, vendor components, and product components susceptible to the vulnerability and at risk of exploitation.

3. The system of claim 1 , wherein the hacker discussion is derived from discussion data retrieved from darkweb based marketplaces and forums.

4. The system of claim 1 , wherein the input data is sorted according to time.

5. The system of claim 1 , wherein the processor executes the machine learning model to implement machine learning techniques with defeasible argumentation to reduce a set of possible labels for each of the set of system components.

6. The system of claim 1 , wherein the machine learning model is constrained to select a set of components related to the system from a reduced platform, a vendor, and a product set.

7. The system of claim 1 , wherein the processor implements an argumentation model configured to construct arguments for a given query that provides arguments supporting decisions indicating why a particular system was identified prior to the cyber-attack from the input data over other systems to support understanding of the identification of the system.

8. The system of claim 1 , wherein the machine learning model uses a text-based feature extracted from data associated with the hacker discussion.

Assignees

• Univ Arizona State
• Securin Inc

Inventors

• Nunes Eric
• Shakarian Jana
• Shakarian Paulo
• Almukaynizi Mohammed
• Singh Harshdeep
• Simari Gerardo
• Sharma Anant

Classifications

• G06F2221/034

  Test or assess a computer or a system · CPC title

• G06N20/00

  Machine learning · CPC title

• G06F16/9536

  Search customisation based on social or collaborative filtering · CPC title

• G06F16/951

  Indexing; Web crawling techniques · CPC title

• G06N7/01

  Probabilistic graphical models, e.g. probabilistic networks · CPC title