Cybersecurity incident response and security operation system employing playbook generation through custom machine learning
A new cybersecurity incident is registered at a security incident response platform. At a playbook generation system, details are received of the new cybersecurity incident from the security incident response platform. At least some of the details correspond to a set of features of the new cybersecurity incident. A set or subset of nearest neighbors of the new cybersecurity incident is localized in a feature space. The nearest neighbors of the new cybersecurity incident are other cybersecurity incidents having a distance from the new cybersecurity incident within the feature space that is defined by differences in features of the nearest neighbors with respect to the set of features of the new cybersecurity incident. A custom playbook is created for responding to the new cybersecurity incident having prescriptive procedures based on occurrences of prescriptive procedures previously employed in response to the nearest neighbor cybersecurity incidents.
1. A computer-implemented method comprising:
encoding values associated with features of a plurality of cybersecurity incidents to obtain a feature vector for each cybersecurity incident, each cybersecurity incident having an associated playbook from a plurality of playbooks for responding to the cybersecurity incident;
receiving details for a new cybersecurity incident;
obtaining a new feature vector for the new cybersecurity incident;
selecting at least one from the plurality of cybersecurity incidents based on the new feature vector, wherein selecting the at least one from the plurality of cybersecurity incidents comprises:
calculating a distance from the new feature vector to each feature vector from the plurality of cybersecurity incidents; and
selecting one or more cybersecurity incidents that are within a predetermined distance from the new feature vector;
determining a score for at least one action from the selected cybersecurity incidents;
selecting one or more actions based on the score of each action;
creating a custom playbook for the new cybersecurity incident that includes the selected one or more actions; and
causing presentation on a display of the custom playbook for responding to the new cybersecurity incident.
2. The method as recited in claim 1 , wherein each playbook comprises at least one action for resolving the associated cybersecurity incident or for completing an investigation of the associated cybersecurity incident.
3. The method as recited in claim 1 , wherein determining the score comprises:
calculating the score for the action as an average relevance of the action weighted by a commitment frequency of the action.
4. The method as recited in claim 1 , wherein the features correspond to incident fields, wherein each incident field is configurable with a corresponding weight.
5. The method as recited in claim 1 , further comprising:
detecting a selection of a user of a new action to respond to the new cybersecurity incident;
adding the new action to the custom playbook; and
adding the new action to at least one from the plurality of cybersecurity incidents.
6. The method as recited in claim 1 , wherein determining the custom playbook for the new cybersecurity incident comprises:
including, in the custom playbook, actions with a score above a predetermined threshold.
7. The method as recited in claim 6 , further comprising:
providing a user interface (UI) to configure one or more thresholds for selecting actions to be added to the custom playbook.
8. The method as recited in claim 7 , further comprising:
detecting that multiple thresholds have been configured; and
presenting in the UI multiple playbooks, each of the multiple playbooks associated with one from the multiple thresholds.
9. A system comprising:
a memory comprising instructions; and
one or more computer processors, wherein the instructions, when executed by the one or more computer processors, cause the system to perform operations comprising:
encoding values associated with features of a plurality of cybersecurity incidents to obtain a feature vector for each cybersecurity incident, each cybersecurity incident having an associated playbook from a plurality of playbooks for responding to the cybersecurity incident;
receiving details for a new cybersecurity incident;
obtaining a new feature vector for the new cybersecurity incident;
selecting at least one from the plurality of cybersecurity incidents based on the new feature vector, wherein selecting the at least one from the plurality of cybersecurity incidents comprises:
calculating a distance from the new feature vector to each feature vector from the plurality of cybersecurity incidents; and
selecting one or more cybersecurity incidents that are within a predetermined distance from the new feature vector;
determining a score for at least one action from the selected cybersecurity incidents;
selecting one or more actions based on the score of each action;
creating a custom playbook for the new cybersecurity incident that includes the selected one or more actions; and
causing presentation on a display of the custom playbook for responding to the new cybersecurity incident.
10. The system as recited in claim 9 , wherein each playbook comprises at least one action for resolving the associated cybersecurity incident or for completing an investigation of the associated cybersecurity incident.
11. The system as recited in claim 9 , wherein determining the score comprises:
calculating the score for the action as an average relevance of the action weighted by a commitment frequency of the action.
12. The system as recited in claim 9 , wherein the features correspond to incident fields, wherein each incident field is configurable with a corresponding weight.
13. The system as recited in claim 9 , wherein the instructions further cause the one or more computer processors to perform operations comprising:
detecting a selection of a user of a new action to respond to the new cybersecurity incident;
adding the new action to the custom playbook; and
adding the new action to at least one from the plurality of cybersecurity incidents.
14. A non-transitory machine-readable storage medium including instructions that, when executed by a machine, cause the machine to perform operations comprising:
encoding values associated with features of a plurality of cybersecurity incidents to obtain a feature vector for each cybersecurity incident, each cybersecurity incident having an associated playbook from a plurality of playbooks for responding to the cybersecurity incident;
receiving details for a new cybersecurity incident;
obtaining a new feature vector for the new cybersecurity incident;
selecting at least one from the plurality of cybersecurity incidents based on the new feature vector, wherein selecting the at least one from the plurality of cybersecurity incidents comprises:
calculating a distance from the new feature vector to each feature vector from the plurality of cybersecurity incidents; and
selecting one or more cybersecurity incidents that are within a predetermined distance from the new feature vector;
determining a score for at least one action from the selected cybersecurity incidents;
selecting one or more actions based on the score of each action;
creating a custom playbook for the new cybersecurity incident that includes the selected one or more actions; and
causing presentation on a display of the custom playbook for responding to the new cybersecurity incident.
15. The non-transitory machine-readable storage medium as recited in claim 14 , wherein each playbook comprises at least one action for resolving the associated cybersecurity incident or for completing an investigation of the associated cybersecurity incident.
16. The non-transitory machine-readable storage medium as recited in claim 14 , wherein determining the score comprises:
calculating the score for the action as an average relevance of the action weighted by a commitment frequency of the action.
17. The non-transitory machine-readable storage medium as recited in claim 14 , wherein the features correspond to incident fields, wherein each incident field is configurable with a corresponding weight.