Model-based system, method, and computer program product for detecting at least potentially unwanted activity associated with confidential data
A model-based system, method, and computer program product are provided for detecting at least potentially unwanted activity associated with confidential data. In use, behavior information associated with use of confidential data is identified, based on predetermined parameters. Additionally, a model is created utilizing the behavioral information. Furthermore, at least potentially unwanted activity associated with the confidential data is detected utilizing the model.
1. A non-transitory computer readable medium comprising instructions that, when executed by at least one processor, cause the at least one processor to at least:
detect an attempt by a user to access confidential data stored on an electronic device;
identify behavioral information based on the confidential data according to one or more predetermined parameters, the predetermined parameters respectively representing a corresponding type of interaction with the confidential data;
create a model utilizing the behavioral information, the model created based on the predetermined parameters and their associated values, the associated values including at least one of an information associated with a predetermined parameter or a weight for the predetermined parameter, the information and weight corresponding to user actions conducted in association with the confidential data;
generate metadata associated with the model, the metadata including an identifier of the user associated with the behavioral information, the identifier including a type of document storing the confidential data;
query a model repository for one or more predefined models, the predefined models generated using metadata profiling prior to the creation of the model;
compare the model with the one or more predefined models representative of known unwanted activity, the unwanted activity including activity disclosing confidential data;
detect at least a first action by the user associated with the confidential data based on the comparison, the first action detected based on a behavioral information threshold, the first action identified as a known unwanted activity when the behavioral information threshold is reached; and
in response to the detection that the first action is the unwanted activity, store the model in a repository as one of the one or more predefined models, the one or more predefined models used to prevent subsequent user access to the confidential data or alert an administrator of the unwanted activity.
2. The non-transitory computer readable medium of claim 1 , wherein the one or more predetermined parameters represent at least one of a scroll of the confidential data, a revisit of the confidential data, a generation of indicia in association with the confidential data, and a digitally-based marking of the confidential data, the digitally-based marking including at least one of highlighting or annotating.
3. The non-transitory computer readable medium of claim 1 , wherein the instructions, when executed, cause the at least one processor to record values reflecting the interaction by the user with the confidential data according to the one or more predetermined parameters.
4. The non-transitory computer readable medium of claim 3 , wherein the instructions, when executed, cause the at least one processor to aggregate the predetermined parameters and the values.
5. The non-transitory computer readable medium of claim 1 , wherein the instructions, when executed, cause the at least one processor to allow the model to be created manually.
6. The non-transitory computer readable medium of claim 1 , wherein the instructions, when executed, cause the at least one processor to create the model automatically.
7. The non-transitory computer readable medium of claim 1 , wherein the instructions, when executed, cause the at least one processor to identify the behavioral information by collecting the behavioral information utilizing an agent.
8. The non-transitory computer readable medium of claim 1 , wherein the instructions, when executed, cause the at least one processor to access the behavioral information via a network.
9. The non-transitory computer readable medium of claim 1 , wherein the identifier of the user includes an identifier of at least one data source associated with the confidential data and a score.
10. A system, comprising:
a processor; and
a memory storing executable instructions that when executed by the processor cause the processor to:
detect an attempt by a user to access confidential data stored on an electronic device;
identify behavioral information based on the confidential data based on one or more predetermined parameters, respective ones of the one or more predetermined parameters corresponding to types of interactions with the confidential data;
create a model utilizing the behavioral information, the model created based on the predetermined parameters and their associated values, the associated values including at least one of an information associated with a predetermined parameter or a weight for the predetermined parameter, the information and weight corresponding to user actions conducted in association with the confidential data;
generate metadata associated with the model, the metadata including an identifier of the user associated with the behavioral information, the identifier including a type of document storing the confidential data;
query a model repository for one or more predefined models, the predefined models generated using metadata profiling prior to the creation of the model;
compare the model with the one or more predefined models representative of known unwanted activity, the unwanted activity including activity disclosing confidential data;
detect at least a first action by the user associated with the confidential data based on the comparison, the first action detected based on a behavioral information threshold, the first action a known unwanted activity when the behavioral information threshold is reached; and
store the model in a repository as one of the one or more predefined models in response to the detection that the first action is the unwanted activity, the one or more predefined models used to prevent subsequent user access to confidential data or alert an administrator of the unwanted activity.