Machine learning

Administrator Help | TRITON AP-DATA | Version 8.3.x

Machine learning classifiers are for more advanced users. In them, you provide examples of the type of data that you want to protect and that you don't want to protect, so the system can learn and identify sensitive data in traffic. These are called positive and negative training sets because the examples educate the system.

Unlike fingerprinting, the files do not need to contain parts of the analyzed files but can look similar or be on a similar topic. The system learns and recognizes complex patterns and relationships and makes decisions on them without exact include/exclude criteria that are specified in fingerprinting classifiers. Machine learning can even protect new, zero-day documents in this way.

Because machine learning classifiers are not looking for an exact match, they can handle a larger number of files.


	Note Machine learning classifiers can be used for unstructured file system data only. They cannot be used for database data or unstructured SharePoint or IBM Domino data.

Once you've created the classifier, the system assesses the expected number of unintended matches (false positives) and undetected content (false negatives) and provides an accuracy level.

The system supports 3 levels of machine learning classifiers:

Explicit negative examples - For example, non-proprietary marketing plans as a negative example to propriety marketing plans.

Non-explicit negative examples- For example, directories that do not contain marketing plans as negative examples to directories with proprietary marketing plan.

Positive examples.

For tips and best practices for using machine learning, refer to Introduction to Machine Learning for TRITON AP-DATA.

Creating a machine learning classifier

Administrator Help | TRITON AP-DATA | Version 8.3.x

Related topics:

Machine Learning Wizard - General

Machine Learning Wizard - Credentials

Machine Learning Wizard - Scanned Folders

Machine Learning Wizard - Scheduler

Machine Learning Wizard - Finish

Machine learning details

To create a machine learning classifier:

Click Main > Policy Management > Content Classifiers.

Select Machine Learning. A list of existing machine learning classifiers appears. You can expand the right pane to view more details, such as last run time, or you can collapse it to show fewer. Click the links in the details pane to adjust classifier settings or view more details. (See Details pane for a description of the details pane.) You can also start, stop, or pause a machine learning process using buttons on the toolbar.

Click New from the menu bar. A wizard opens. There are several pages in the wizard:

Complete the information on each page and click Next to proceed through the wizard.