Systems and methods for using multiple in-line heuristics to reduce false positives
AssigneeSymantec Corporation
InventorMark Kennedy
Life of this patent
Estimating the term…
Abstract
An exemplary method for using multiple in-line heuristics to reduce false positives may include: 1) training a first heuristic using a set of training data, 2) deploying the first heuristic, 3) identifying false positives produced by the first heuristic during deployment, 4) modifying the training data to include the false positives produced by the first heuristic, 5) creating a second heuristic using the modified training data, 6) deploying both the first heuristic and the second heuristic, and then 7) applying both the first heuristic and the second heuristic, in sequence, to a set of field data.
Claims20
A computer-implemented method for using multiple in-line heuristics to reduce false positives, at least a portion of the method being performed by a computing device comprising at least one processor, the method comprising:
- training a first heuristic using a set of training data;
- deploying the first heuristic;
- identifying false positives produced by the first heuristic during deployment;
- modifying the training data to include the false positives produced by the first heuristic;
- creating a second heuristic using the modified training data;
- deploying both the first heuristic and the second heuristic;
- applying both the first heuristic and the second heuristic, in sequence, to a set of field data.
A system for sequentially applying related heuristics to reduce false positives, the system comprising:
- a training module programmed to train a first heuristic using a set of training data;
- a deployment module programmed to deploy the first heuristic;
- an identification module programmed to identify false positives produced by the first heuristic during deployment;
- a tuning module programmed to:
- modify the training data to include the false positives produced by the first heuristic;
- create a second heuristic using the modified training data;
- wherein the deployment module is further programmed to:
- deploy both the first heuristic and the second heuristic;
- apply both the first heuristic and the second heuristic, in sequence, to a set of field data;
- a processor configured to execute the training module, the deployment module, the identification module, and the tuning module.
A computer-readable medium comprising computer-executable instructions that, when executed by at least one processor of a computing device, cause the computing device to:
- train a first heuristic using a set of training data;
- deploy the first heuristic;
- identify false positives produced by the first heuristic during deployment;
- modify the training data to include the false positives produced by the first heuristic;
- create a second heuristic using the modified training data;
- deploy both the first heuristic and the second heuristic;
- apply both the first heuristic and the second heuristic, in sequence, to a set of field data.
Dependencies are inferred from the claim text.