Methods and systems for generating recommendations based on threat model knowledge graphs comprising crowdsourced modeling contributions

Patent metadata

Field	Value
Publication number	US-12184684-B2
Application number	US-202117542185-A
Country	US
Kind code	B2
Filing date	Dec 3, 2021
Priority date	Dec 3, 2021
Publication date	Dec 31, 2024
Grant date	Dec 31, 2024

Abstract

The methods and systems relate to improvements to threat modeling systems through the use of crowdsourcing. Specifically, the methods and systems relate to generating recommendations based on crowdsourced threat modeling contributions. For example, the methods and systems automate the threat modeling process by leveraging data in order to drive consistent and measurable quality of threat models and enable threat models to provide aggregated views of risk concentration at any altitude.

First claim

What is claimed is:

1. A system for generating recommendations based on threat model knowledge graphs comprising crowdsourced modeling contributions, the system comprising:
cloud-based storage circuitry configured to store a threat model knowledge graph, wherein the threat model knowledge graph is a graph-structured data model for integrating threat model contributions;
control circuitry configured to:
receive a first threat model contribution from a first user, wherein the first threat model contribution has a first contribution characteristic, and wherein the first threat model contribution comprises first content for publishing in a first threat modeling system;
determine, based on the first contribution characteristic, a first location in the threat model knowledge graph for the first threat model contribution;
receive a second threat model contribution from a second user, wherein the second threat model contribution has a second contribution characteristic, and wherein the second threat model contribution comprises second content for publishing in the first threat modeling system;
determine, based on the second contribution characteristic, a second location in the threat model knowledge graph for the second threat model contribution;
add the first threat model contribution and the second threat model contribution to the threat model knowledge graph at the first location and the second location, respectively;
in response to adding the first threat model contribution and the second threat model contribution to the threat model knowledge graph, update a threat modeling system based on the threat model knowledge graph;
receive a user request for a recommendation based on the threat modeling system, wherein the recommendation compares the first threat modeling system to a second threat modeling system; and

input/output circuitry configured to:
generate for display, in a user interface, the recommendation.

2. A method of generating recommendations based on threat model knowledge graphs comprising crowdsourced modeling contributions, the method comprising:
receiving a first threat model contribution from a first user, wherein the first threat model contribution has a first contribution characteristic;
determining, based on the first contribution characteristic, a first location in a threat model knowledge graph for the first threat model contribution;
receiving a second threat model contribution from a second user, wherein the second threat model contribution has a second contribution characteristic;
determining, based on the second contribution characteristic, a second location in the threat model knowledge graph for the second threat model contribution;
adding the first threat model contribution and the second threat model contribution to the threat model knowledge graph at the first location and the second location, respectively;
in response to adding the first threat model contribution and the second threat model contribution to the threat model knowledge graph, updating a first threat modeling system based on the threat model knowledge graph; and
generating for display, in a user interface, a recommendation based on the first threat modeling system, wherein the recommendation compares the first threat modeling system to a second threat modeling system.

3. The method of claim 2 , wherein the first contribution characteristic indicates a first software architecture upon which the first threat model contribution is based, and wherein the second contribution characteristic indicates a second software architecture upon which the second threat model contribution is based.

4. The method of claim 2 , wherein the first contribution characteristic indicates a first mitigation technique upon which the first threat model contribution is based, and wherein the second contribution characteristic indicates a second mitigation technique upon which the second threat model contribution is based.

5. The method of claim 2 , wherein the first contribution characteristic indicates a first data flow upon which the first threat model contribution is based, and wherein the second contribution characteristic indicates a second data flow upon which the second threat model contribution is based.

6. The method of claim 2 , wherein determining, based on the first contribution characteristic, the first location, further comprises:
retrieving a first contribution characteristic keyword for the first contribution characteristic;
retrieving a second contribution characteristic keyword for available locations in the threat model knowledge graph;
determining a first similarity between the first contribution characteristic keyword and the second contribution characteristic keyword;
comparing the first similarity to a threshold similarity; and
determining that the first contribution characteristic keyword and the second contribution characteristic keyword correspond based on the first similarity equaling or exceeding the threshold similarity.

7. The method of claim 2 , further comprising:
determining a third location in the threat model knowledge graph that is unpopulated;
retrieving a third contribution characteristic for the third location;
identifying a third user corresponding to the third contribution characteristic; and
transmitting a request to the third user for a third threat model contribution for adding to the threat model knowledge graph at the third location.

8. The method of claim 2 , further comprising:
receiving a third threat model contribution, wherein the third threat model contribution has a third contribution characteristic;
inputting the third threat model contribution into a machine learning model, wherein the machine learning model is trained to selection locations for adding threat model contributions to the threat model knowledge graph; and
receiving an output of the machine learning model indicating a third location for adding the third threat model contribution to the threat model knowledge graph.

9. The method of claim 2 , wherein updating the first threat modeling system comprises:
receiving a first user characteristic for the first user;
comparing the first user characteristic to user characteristics required for updating the first threat modeling system; and
validating the first threat model contribution at the first location based on the first user characteristic corresponding to user characteristics required for updating the first threat modeling system.

10. The method of claim 2 , wherein updating the first threat modeling system comprises:
generating for display, in the user interface, a request to add the first threat model contribution to the threat model knowledge graph;
receiving a number of user confirmations for adding the first threat model contribution to the threat model knowledge graph;
comparing the number of user confirmations to a threshold number; and
validating the first threat model contribution based on the number of user confirmations corresponding to the threshold number.

11. The method of claim 2 , wherein the first threat model contribution comprises first content for publishing in the first threat modeling system, and wherein the second threat model contribution comprises second content for publishing in the first threat modeling system.

12. A non-transitory, computer-readable medium for generating recommendations based on threat model knowledge graphs comprising crowdsourced modeling contributions, comprising instructions that, when executed by one or more processors, cause operations comprising:
receiving a first threat model contribution from a first user, wherein the first threat model contribution ha

Assignees

• Capital One Services Llc

Inventors

• Underwood Jonathan
• Barlow Neil
• Scott Fraser Richard

Classifications

• H04L41/14

  Network analysis or design · CPC title

• H04L63/1441 Primary

  Countermeasures against malicious traffic (countermeasures against attacks on cryptographic mechanisms H04L9/002) · CPC title

• H04L41/16

  using machine learning or artificial intelligence · CPC title

• H04L63/20

  for managing network security; network security policies in general (filtering policies H04L63/0227) · CPC title

• G06F21/577

  Assessing vulnerabilities and evaluating computer system security · CPC title