Abnormal cross authorization detection systems
A system to detect abnormal cross authorizations and take action is described. The system determines whether cross authorization event applied to a first trained anomaly detection model and activity post cross authorization event applied to a second trained anomaly detection model is suspicious. An indicator score is determined from rule-based security indications applied to the cross authorization. A security action is taken based on application of the indicator score applied to a threshold.
1. A method to provide cybersecurity, the method comprising:
determining whether a cross authorization event applied to a first trained anomaly detection model and activity post cross authorization event applied to a second trained anomaly detection model are suspicious, wherein at least one of the following:
the first trained anomaly detection model employs a first plurality of feature vectors comprising first aspects from a perspective of a first management system and a second management system; or
the second trained anomaly detection model employs a second plurality of feature vectors comprising second aspects from the perspective of the first management system and the second management system;
determining an indicator score from rule-based security indications applied to the cross authorization event; and
taking a security action based on application of the indicator score to a threshold.
2. The method of claim 1 , wherein the first trained anomaly detection model employs the first plurality of feature vectors comprising the first aspects from the perspective of the first management system and the second management system.
3. The method of claim 1 , wherein the second trained anomaly detection model employs the second plurality of feature vectors comprising the second aspects from the perspective of the first management system and the second management system.
4. The method of claim 1 , wherein the determining whether the cross authorization event is suspicious comprises:
generating a model security score from the first trained anomaly detection model and the second trained anomaly detection model.
5. The method of claim 4 , wherein the model security score is applied to a selected threshold.
6. The method of claim 5 , wherein the model security score is based on a combination of an access elevation anomaly security score generated by the first trained anomaly detection model and a post elevation activity security score generated by the second trained anomaly detection model.
7. The method of claim 5 , wherein the indicator score is determined based on the model security score exceeding the selected threshold.
8. The method of claim 1 , wherein taking the security action comprises:
issuing a security alert.
9. The method of claim 1 , wherein at least one of the following:
the first trained anomaly detection model employs the first plurality of feature vectors comprising the first aspects from the perspective of an identity and access management system and a resource management system; or
the second trained anomaly detection model employs the second plurality of feature vectors comprising the second aspects from the perspective of the identity and access management system and the resource management system.
10. A computer readable storage device to store computer executable instructions to control a processor to:
determine whether a cross authorization event applied to a first trained anomaly detection model and activity post cross authorization event applied to a second trained anomaly detection model are suspicious, wherein at least one of the following:
the first trained anomaly detection model generates a first plurality of feature vectors corresponding to a plurality of perspectives of a plurality of management systems; or
the second trained anomaly detection model generates a second plurality of feature vectors corresponding to the plurality of perspectives of the plurality of management systems;
determine an indicator score from rule-based security indications applied to the cross authorization event; and
take a security action based on application of the indicator score to a threshold.
11. The computer readable storage device of claim 10 , wherein the first trained anomaly detection model generates the first plurality of feature vectors corresponding to the plurality of perspectives of the plurality of management systems.
12. The computer readable storage device of claim 10 , wherein the second trained anomaly detection model generates the second plurality of feature vectors corresponding to the plurality of perspectives of the plurality of management systems.
13. The computer readable storage device of claim 10 , wherein at least one of the following:
the first plurality of feature vectors comprises a first resource access management perspective feature vector and a first identity and access management perspective feature vector; or
the second plurality of feature vectors comprises a second resource access management perspective feature vector and a second identity and access management perspective feature vector.
14. A system, comprising:
a memory device to store a set of instructions; and
a processor to execute the set of instructions to:
determine whether a cross authorization event applied to a first trained anomaly detection model and activity post cross authorization event applied to a second trained anomaly detection model are suspicious;
determine an indicator score from rule-based security indications applied to the cross authorization event; and
take a security action based on application of the indicator score to a threshold, wherein the system is in a cloud-based environment that comprises at least one of an identity and access management system or a resource management system.
15. The system of claim 14 , wherein the cloud-based environment comprises the identity and access management system.
16. The system of claim 15 , wherein the first trained anomaly detection model employs a plurality of feature vectors comprising aspects from a perspective of the identity and access management system.
17. The system of claim 14 , wherein the cloud-based environment comprises the resource management system.
18. The system of claim 17 , wherein the second trained anomaly detection model employs a plurality of feature vectors comprising aspects from a perspective of the resource management system.
19. The system of claim 14 , wherein the cloud-based environment comprises the identity and access management system and the resource management system.
20. The system of claim 19 , wherein the first trained anomaly detection model employs a first plurality of feature vectors comprising first aspects from a perspective of the identity and access management system and the resource management system; and
wherein the second trained anomaly detection model employs a second plurality of feature vectors comprising second aspects from the perspective of the identity and access management system and the resource management system.