Detecting malicious beaconing communities using lockstep detection and co-occurrence graph

1. A cloud-implemented method, comprising:

providing, as input data, information of a listing of potential beacon incidents as output data from a beaconing detector;

reformatting the input data into bipartite graph data in a format of source MAC (Media Access Control) address data versus destination IP (Internet Protocol) data and timestamp information;

providing the bipartite graph data as input data into a first processing to detect malicious beaconing activities using a lockstep detection module on the input bipartite graph data, as executed in a cloud environment, to detect possible synchronized attacks against a targeted infrastructure; and

providing the bipartite graph data as input data into a second processing to provide community detection by a community detection module of the second processing for processing a community detection mechanism that evaluates a domain server churn by evaluating edge weights on co-occurrence graph data from a co-occurrence graph conversion module.

2. The cloud-implemented method of claim 1 , further comprising:

merging results of the first processing and results of the second processing; and

providing the merged results as output data from an output module.

3. The cloud-implemented method of claim 1 , wherein the first processing further comprises:

extracting bipartite graphs at different time windows;

building a frequency pattern tree, wherein the frequency pattern tree includes temporal information; and

detecting lockstep patterns from the frequency pattern tree.

4. The cloud-implemented method of claim 1 , wherein the first processing further comprises detecting communities based on at least edge weights of the projection graph data,

further comprising providing the input bipartite graph data into the second processing executed in a cloud environment, the second processing initially converting the bipartite graph data into a co-occurrence graph format that indicates in a graph format how devices in a targeted infrastructure communicate with different external destination servers over time, the second processing detecting malicious beaconing activities by analyzing data exchanges with the external destination servers over a plurality of time windows to detect anomalies, using the data in the co-occurrence graph format.

5. The cloud-implemented method of claim 1 , wherein the first processing further comprises:

filtering out destinations based on a comparison with a white-list; and

providing the filtered destination information as an output of the first processing.

6. The cloud-implemented method of claim 1 , wherein the second processing comprises extracting features from egonets of destinations from data in the co-occurrence graph format.

7. An apparatus for cloud-based implementation, comprising:

a processor; and

a memory device,

wherein the memory device stores a set of instruction permitting the processor to execute a cloud-implemented method of:

providing, as input data, information of a listing of potential beacon incidents as output data from a beaconing detector;

reformatting the input data into bipartite graph data in a format of source MAC (Media Access Control) address data versus destination IP (Internet Protocol) data and timestamp information;

providing the bipartite graph data as input data into a first processing to detect malicious beaconing activities using a lockstep detection module on the input bipartite graph data, as executed in a cloud environment, to detect possible synchronized attacks against a targeted infrastructure; and

providing the bipartite graph data as input data into a second processing to provide community detection by a community detection module of the second processing for processing a community detection mechanism that evaluates a domain server churn by evaluating edge weights on co-occurrence graph data from a co-occurrence graph conversion module.

8. The apparatus of claim 7 , wherein the processor is further configured to:

merge results of the first processing and results of the second processing; and

provide the merged results as output data from an output module.

9. The apparatus of claim 7 , wherein the first processing further comprises:

extracting bipartite graphs at different time windows;

building a frequency pattern tree, wherein the frequency pattern tree includes temporal information; and

detecting lockstep patterns from the frequency pattern tree.

10. The apparatus of claim 7 , wherein the first processing further comprises detecting communities based on at least edge weights of the projection graph data,

further comprising providing the input bipartite graph data into the second processing executed in a cloud environment, the second processing initially converting the bipartite graph data into a co-occurrence graph format that indicates in a graph format how devices in a targeted infrastructure communicate with different external destination servers over time, the second processing detecting malicious beaconing activities by analyzing data exchanges with the external destination servers over a plurality of time windows to detect anomalies, using the data in the co-occurrence graph format.

11. The apparatus of claim 7 , wherein the first processing further comprises:

filtering out destinations based on a comparison with a white-list; and

providing the filtered destination information as an output of the first processing.

12. The apparatus of claim 7 , wherein the second processing comprises extracting features from egonets of destinations from data in the co-occurrence graph format.

13. A computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions readable and executable by a computer to cause the computer to perform a method in a cloud-based implementation, comprising:

providing, as input data, information of a listing of potential beacon incidents as output data from a beaconing detector;

reformatting the input data into bipartite graph data in a format of source MAC (Media Access Control) address data versus destination IP (Internet Protocol) data and timestamp information;

providing the bipartite graph data as input data into a first processing to detect malicious beaconing activities using a lockstep detection module on the input bipartite graph data, as executed by a processor on a computer, to detect possible synchronized attacks against a targeted infrastructure; and

providing the bipartite graph data as input data into a second processing to provide community detection by a community detection module of the second processing for processing a community detection mechanism that evaluates a domain server churn by evaluating edge weights on co-occurrence graph data from a co-occurrence graph conversion module.

14. The computer program product of claim 13 , further comprising:

merging results of the first processing and results of the second processing; and

providing the merged results as output data from an output module.

15. The computer program product of claim 13 , wherein the first processing further comprises:

extracting bipartite graphs at different time windows;

building a frequency pattern tree, wherein the frequency pattern tree includes temporal information; and

detecting lockstep patterns from the frequency pattern tree.

16. The computer program product of claim 13 , wherein the first processing further comprises detecting communities based on at least edge weights of the projection graph data,

further comprising providing the input bipartite graph data into the second processing executed in a cloud environment, the second processing initially converting the bipartite graph data into a co-occurrence graph format that indicates in a graph format how devices in a targeted infrastructure communicate with different external destination servers over time, the second processing detecting malicious beaconing activities by analyzing data exchanges with the external destination servers over a plurality of time windows to detect anomalies, using the data in the co-occurrence graph format.

17. The computer program product of claim 13 , wherein the first processing further comprises:

filtering out destinations based on a comparison with a white-list; and

providing the filtered destination information as an output of the first processing,

wherein the second processing comprises extracting features from egonets of destinations from data in the co-occurrence graph format.